LIBRARY OF THE UNIVERSITY OF CALIFORNIA. Class \A7 OP^S Jay fr) QUALITATIVE CHEMICAL ANALYSIS. A Guide in Qualitative Work, with data for Analytical Operations and Laboratory Methods in Inorganic Chemistry, by ALBERT B. PRESCOTT and OTIS C. JOHNSON, Professors in the University of Michigan. Fifth revised and enlarged edition, entirely rewritten. 8vo, cloth $3. SO Net. OUTLINES OF PROXIMATE ORGANIC ANALY- SIS : For the Identification, Separation, and Quantitative Determination of the more Com- monly Occurring Organic Compounds. Fourth edition. 12mo, cloth $1.75 FIRST BOOK IN QUALITATIVE CHEMISTRY. Tenth edition, revised. 12mo, cloth $1.5O ORGANIC ANALYSIS: A MANUAL OF THE DESCRIPTIVE AND ANALYTICAL CHEMISTRY OF CERTAIN CARBON COMPOUNDS IN COMMON USE. QUALITATIVE AND QUANTITATIVE ANALYSIS OF ORGANIC MATERIALS: COMMERCIAL AND PHARMACEUTICAL ASSAYS; THE ESTIMATION OF IMPURITIES UNDER AUTHORIZED STANDARDS ; FORENSIC EXAMINATIONS FOR POISONS ; AND ELEMENTARY ORGANIC ANALYSIS. BY ALBERT B. PRESCOTT, PH.D., M.D., Director of the Chemical Laboratory in the University of Michigan, Author of " Outlines of Proximate Organic Analysis," " Qualitative Chemical Analysis" etc. FIFTH EDITION. NEW YORK: D. VAN NOSTRAND COMPANY, 23 MURRAY A1TD 27 WARREN STREET. 1901. COPTBIGHT, 1887, BY W. H. FAKRINGTON. GENERAL PREFACE. THE operator in chemical analysis requires for his direction a system of descriptive chemistry, to be as nearly complete as possible. In resorting to the hand-books of general chemistry for the record of physical and chemical constants the analyst is often disappointed. It belongs, therefore, to analytical chemistry to furnish chemical descriptions with special precision, and this is a service promoting independent chemical work. As a mere changeful body of directions, giving the latest expedients in methods, analytical chemistry cannot claim to have educational value. But as an operative introduction to the character and deportment of compounds, analysis becomes a logical mode of study, fruitful of important results. For certain common carbon compounds it has been under- taken to furnish in this work, first, systematic chemical description, and thereupon the methods of analytical procedure, qualitative, quantitative, and for proofs of purity, all with liberal citations of the authorities for convenience of further reading. In the references an order is observed as follows : (1) name of the con- tributor, (2) year of the contribution, (3) volume and page, first of original and then of contemporary publications. Respecting the assumed peculiarities of organic analysis, it more and more appears that the differences between inorganic and organic analysis have been greatly overstated, just as, at earlier periods, the distinction between inorganic and organic chemistry in general was overdrawn. With nearer acquaintance it is seen that the limits of error in determination of carbon compounds are by no means always wider than those in analysis of metallic bodies. vi PREFACE, If the author of this work have done anything at all to rescue the analytical chemistry of carbon compounds from a disjointed position in chemical literature, he will have gained enough of recompense. He desires to make* thankful acknowledgment of the encouraging favor which has been extended to his " Outlines of Proximate Organic Analysis" since its issue in 1874. While his own promise of further publication has waited long for ful- filment, works of distinct value have opportunely appeared in different parts of the same field, and the flow of good contribu- tions has continued to increase everywhere. Organic analysis, as the determination of the unbroken compounds of carbon, no longer has an uncertain place in chemical learning. ALBERT B. PKESCOTT. UNIVERSITY OF MICHIGAN, ANN ARBOR, October, 1887. ORGANIC ANALYSIS. ABSINTHIN. C 20 H 28 O 4 . H 2 O=350. The neutral princi- ple of the wormwood, Artemisia absinthium. Obtained by pre- cipitating the hot-water extract of the leaves and tops by tannic acid, drying the precipitate with litharge, and extracting with alcohol. The absinthin may be purified by filtering the alcoholic solution through animal charcoal, evaporating, and redissolving in ether. Absinthin solidifies from yellow drops to indistinct crystals, melting at 120 C., and decomposing at higher temperatures. It has an aromatic odor and a very bitter taste. It is almost in- soluble in cold water, slightly soluble in hot water, freely soluble in alcohol or ether ; soluble with a brown-red color in the alkali hydrates. The potassa solution, when acidified by hydro- chloric acid, exhibits a yellow-green play of colors. Concentrat- ed sulphuric acid dissolves it with brown color changing to green-blue, and becoming dark blue on adding a very little water. Much water decolors it. If the alcoholic solution be treated with an equal volume of concentrated sulphuric acid a brown - red mixture results, and a violet color is obtained after adding a few drops of water. Froehde's reagent gives a brown color changing to green and violet (BACH, 1874). Absinthin precipi- tates mercurous nitrate dirty-yellow ; lead subacetate brown - yellow ; barium acetate brown. Soiling with dilute acids de- composes absinthin without producing a glucose. Fehling's so- lution is not reduced by it : ammoniacal s'ilver nitrate solution is reduced, with the formation of a mirror. ACETIC ACID. Essigsaure. Acide acetique. C 2 H 4 O 2 = 60. Methyl-carboxyl, CH 3 . CO 2 H. Manufactured from alcohol or dilute alcoholic liquids by oxidation, or the acetous " fermen- tation," and from wood by destructive distillation yielding other products of value. It is produced in numerous chemical re- actions. Acetic acid is identified by its odor in the free state (b] and the more intense odor of its ethyl ester (d). The empyreuma of 8 ACETIC ACID. heated acetates is characteristic (d). It gives a distinctive color with ferric salts (d). It is separated by distillation, if necessary preceded by saponification (e). From buty rates, by the insolu- bility of the barium acetate in alcohol (Butyric acid, e). It is estimated, as free acid, by acidimetry (f), or gravimetric satura- tion ; as alkali acetate, by the alkalimetry of the ignited residue (f). In Acetate of Lime by distillation and by special methods (p. 11). Commercial Grades and Impurities, p. 14. Vinegar, its standards of strength, impurities, and special tests, p. 15. a. Absolute acetic acid (Glacial Acetic acid, Eisessig) below about 15 C. is a crystalline solid, forming transparent tabular masses, melting at 16.7 C. to a colorless liquid* An acid of 87 per cent, melts below C. ; of 62 per cent, at 24 C. The absolute acid boils at 118 C. It has, at 15 C., the sp. gr. 1.0607 (water at 4 C.) (MENDELEJEFF). b. Acetic acid has a pure acidulous taste and a penetrating, vinegar-like odor. When concentrated it is an irritant to the skin or tongue, and should be diluted before tasting. c. Acetic acid is soluble in all proportions of water and alcohol ; the absolute acid is soluble in all proportions of ether, and acts as a solvent for various essential oils, resins, camphors, phenols, and metallic salts. Diluted with water acetic acid gives an acid reaction with litmus. The metallic normal acetates are soluble in water ; silver and mercurous acetates less freely than the others. Perfectly normal alkali acetates are neutral in re- action, a*s shown by phenol-phthalein or litmus, but potassium acetate is liable to be found alkaline, because slightly basic. Acetates in general lose acetic acid in hot solution, and in some instances by simple exposure, so that acetates exhale a percep- tible acetous odor, and gradually become basic, o^on-alkali ace- tates, in solution, become slightly turbid, by formation of car- bonate, from carbon dioxide of the air. d. Ferric chloride or other ferric salt, added not in ex- cess to solution of acetates, causes a red color by formation of ferric acetate. On boiling, a yellow-brown precipitate of basic acetate of iron is obtained, resolved finally into nearly pure ferric hydrate. The red liquid, before heating, is not decolored by adding mercuric chloride solution, nor taken up by shaking with ether, both these negative results giving distinction from Thio- cyanic acid. The color is destroyed by adding sulphuric or hydrochloric acid a distinction from Meconic acid. By hot digestion with sulphuric acid and alcohol, ethyl acetate, or ACETIC ACID. 9 acetic ether, is formed, recognized by its penetrating, fragrant odor. This test is most efficient when the dry acetate, obtained from acidulous liquid by neutralizing with fixed alkali and eva- porating, is treated with an equal quantity of alcohol and a double quantity of sulphuric acid, and heated or distilled. The odor of other ethyl esters is liable to be mistaken for this. When dry acetates are strongly heated in a test-tube, carbon is separated and acetone, CgHgO, is evolved, capable of recognition by its odor. By distillation of acetates with phosphoric or sulphuric acid, free acetic acid is obtained, with its characteristic odor. Acetic acid is a stable compound, not oxidized by chromic acid nor by permanganates. c e. Separations. Aqueous solutions of acetates, if kept slightly alkaline with fixed alkali, can be concentrated without loss of acetic acid. The free acid distils very slowly, and its quantitative distillation requires thorough treatment. In distil- ling from acetates, phosphoric or sulphuric acid, or oxalic aci$, is to be added, in some excess of the quantity needful to form nor- mal salts with all the bases present. To obtain all the acetic acid it is necessary to distil to dryness, adding water and repeating several times, until the distillate ceases to be acid to litmus. When various organic matters are present, it is therefore usually better to displace with phosphoric acid, avoiding the action of sulphuric acid in distilling to dryness. Care is to be taken that the phosphoric acid is strictly free from volatile acids, and that salts of volatile acids other than acetic are not present. If hy- drochloric acid or its salts are present, the addition of sufficient silver sulphate insures the retention of the chlorine. Further details respecting quantitative distillation are given under f. To obtain the acetic acid of basic acetates insoluble in water, it is preferable to transpose them to alkali acetate by digesting with hot solution of sodium carbonate, filtering, and exhausting with hot water. The same operation may with advantage pre- cede distillation in the case of lead acetate. Ethereal acetates, such as ethyl acetate, do not give up their acetic acid by displac- ing it with a non-volatile acid, but require first to be saponified by an alkali, when the alkali acetate is treated as before de- scribed. The saponification is effected by digesting with some excess of a solution of potassa in alcohol free from acetic acid, when all the alcohol may be removed by evaporation. Also, a volumetric estimation of the acetic acid of ethereal acetates may be readily and exactly made by saponifying with a known quan- tity of alcoholic potassa (see/"). io ACETIC ACID. f. Quantitative. In simple dilution with water, the spe- cific gravity of acetic acid, if closely taken, is a practicable indi- cation of percentage, according to tables of accepted authority, bearing in mind that acid of about 46 per cent, coincides in density with acid of 99 per cent. Even within the range to which it applies, the hydrometer is not exact enough, unless cor- rected in its reading by the analyst himself. Saturation methods of estimation are to be preferred, especially that by voliimetric solution of fixed alkali. Phenol-phthalein is the best indicator, but litmus will serve. Colored liquids may be diluted so as to show the phenol phthalein indication. If 6.000 grams of the acid mixture jbe taken, each c.c. of normal solution of alkali indi- cates 1 per cent, of C 2 H 4 O 2 , or real acid ; each c c. of decinor- mal alkali, 0.1 per cent. With dilute acetic acid, 24.0 grams may be taken, when c.c. -f- 4 = %. But, owing to the vaporization of acetic acid, it is seldom advisable to take a stated weight for esti- mation. In a stoppered bottle, previously tared, pour 5 to 6 c.c. of the acid under estimation, stopper, take the weight, and titrate ; grams taken : 6.000 :: c.c. of normal alkali : a? = per cent, real acid. Gravimetric methods of saturation may be employed. 1.000 tram of potassium 'bicarbonate (or 0.530 gram dry sodium car- onate), taken in a tall beaker, may be neutralized with the acetic acid, the acid being added by weight from a small, light, lipped beaker, carrying a small glass rod with which to pour, adding at last drop by drop, and heating to expel the carbon dioxide. Then 60 -f- grams of acid required = number per cent, of real acid present. Before testing the acetic acid, if much stronger than vinegar, it should be diluted, by weight, to from 2 to 15 times its own weight, so. as not to be over 5 to Si strength. Then 60 X the factor of dilution (2 to 15)-r-nurnber grams of the diluted acid required = per cent, of real acid present. A gravi- metric method with barium carbonate is as follows : A weighed quantity of the acetic acid (sufficient to contain 0.120 to 0.180 gram absolute acetic acid) is digested with excess of well-washed, precipitated barium carbonate, the precipitate is filtered and ex- hausted with hot water, the filtrate is precipitated by dilute sul- phuric acid, with heating and washing as required in estimation of barium as sulphate, and the ignited barium sulphate weighed. (BaSO 4 : 2C 2 H 4 O 2 : : 232.8 : 120 : : 1 : 0.5156.) Grams of barium sulphate X 0.5156 = grams acetic acid absolute, in the quantity of acetic acid mixture under estimation. Free acids which form insoluble barium salts do not interfere. Oxalic acid will add by a trifling quantity to the result. Free acids which form soluble ACETATE OF LIME. II barium salts interfere altogether, but the addition of sufficient silver sulphate prevents interference of hydrochloric acid. Ace- tates and other salts of non-alkali metals precipitable by barium carbonate cannot be present. The acetic acid of alkali and alkaline earth salts may bo estimated by ignition of the dry salt, and titration of the result- ing alkali carbonate, or alkaline earth, with volumetric acid. Each c.c. of normal solution of acid used indicates 0.06 gram of absolute acetic acid. Of course the acetate taken for estimation in this way must be of neutral reaction ; or, if of alkaline reac- tion, its alkalinity (before ignition) must be estimated by titration, and the c.c. of acid so used must be deducted from the c.c. re- quired in titrating the ignited residue from an equal quantity of the salt. This plan of estimation is not among the more trust- worthy ones. The acetic acid of normal acetates of calcium, lead, and other non-alkali metals, is sometimes estimated by methods of determination of the metal. Valuation of "Acetate of Lime" Acetate of Lime (Pyro- lignate of Lime, Essigsauren Kalk, Holzessigsauren Kalk) is a product of the distillation of wood, used as a carrier of acetic acid toward concentration and purification. Its value lies in the amount of real acetic acid it contains. Three grades of it have been made the a gray," "brown. "and " black " but the last- txained grade is now seldom produced. Besides empyreumatic a. id carbonaceous matters, it is quite liable to contain butyrate, formate, and propionate ; 1 also magnesium salts ; and may con- tain chlorides. In the plan of wood distillation conducted at temperatures below charring, 2 Acetate of Sodium is usually manufactured instead of lime acetate, and no empyreumatic matter occurs. In the valuation of acetate of lime, the methods mostly in use have been based on (1) distillation of the acetic acid, and (2) the amount of soluble lime salts present. A volu- metric method (3) with evaporation of the acetic acid will also be given here. 3 The valuation should embrace an estimation of the moisture, and may present the proportion of magnesium ace- tate, if any be present. Samples are to be taken from every 1 Respecting the relation of these impurities to methods of estimation, LUCK, 1871 : Zeitsch. anal Chem., 10, 184. 2 MABERY, 1883 : Am. Chem. Jour., 5, 256. 3 Respecting methods (1) and (2) STILLWELL and GLADDING, 1882 : Jour. Amtr. Chem. Soc.. 4, 94. SEELY, 1872 : Am. Chem., 2, 324 ; 3, 8. FRESE- NIUS, 1875 : Zeitsch. anal Chem., 14, 172 ; 1866 : Ibid., 5, 315 ; 1874 : Ibid., 13, 153. H. ENDEMANN, 1876 : Am. Chem., 6, 294. A. A. BLAIR, 1885 : Am. Chem. Jour., 7, 26. 12 ACETIC ACID. fifth to tenth bag, fairly representing both large and small pieces, and inclosed in rubber bags or air-tight jars while sent and held for analysis. The moisture is always to be determined in a portion taken as soon as the sample is opened to the air. The sample is then pulverized and sifted in preparation for the ana- lysis. Then a prepared portion taken parallel with that sub- jected to analysis is dried for estimation of its moisture, from which the percentage of acetic acid is at last corrected for mois- ture, whether for the figures on a dry basis, or on the air-dry basis of the primary samples (Stillwell and Gladding). Crystal- lized acetate of calcium contains water of crystallization and is efflorescent ; the product " acetate of lime " may gain or lose water in the air^ but in paper or wood packages it is likely to lose. (1) By distillation of the acetic acid. The most trustworthy method. Of the prepared sample 5 grams are dissolved in 50 c.c. of water, at least 25 grams of glacial phosphoric acid are added, and the liquid distilled, repeatedly adding water, not per- mitting the liquid to be reduced to dryness, and persisting until the distillate ceases to have an acid reaction, or the retort to smell of acetic acid. According to Messrs. Stillwell and Glad- ding, if the retorted liquid be not reduced too low, not more than traces of hydrochloric acid can be carried over from chlo- rides, and the excess of phosphoric acid prevents production of insoluble calcium phosphate. All distillation of hydrochloric acid can be prevented by adding silver sulphate in the retort. Nitric acid must be tested for. Fresenius (1875) and Endemann (1876) describe apparatus by which steam is introduced into the retort, in a current of regulated force, for continuing the distilla- tion. The total distillate is made to a desired definite volume, an aliquot part is measured out, phenol-phthalein added as an indicator, and titrated with standard solution of alkali (p. 10). (2) Methods depending on the quantity of soluble lime salts present. Of these methods the one given by Fresenius (1874, where cited) is one of the best, and is adapted to the assay of pure grades, free from acid empyreuma and from magnesium salt. Of the sample 5 grams are treated with about 150 c.c. of water in a quarter-liter flask, 70 to 80 c.c. -of normal solution of oxalic acid added, and the mixture diluted with water to the 250 c.c. mark. To compensate for the volume of the precipitate 2.1 c c. of water are added above the mark. After being shaken and standing for some time the precipitate is filtered out (through a dry filter). Of the filtrate 100 c.c. are titrated with normal ACETATE OF- LIME. 13 solution of alkali for acid as acetic acid. Then another portion of 100 c.c. is treated with calcium acetate to precipitate all the excess of oxalic acid. The calcium oxalate precipitate is filtered out, washed, dried, ignited, weighed as calcium carbonate, and the indicated quantity of oxalic acid calculated into its equiva- lent of acetic acid. The total acid as acetic acid in ] 00 c.c., minus the oxalic acid as acetic acid in 100 c.c., equals the true acetic acid in 100 c.c. of filtrate that is, in f of the sample as- sayed (or in 2 grams). (3) A method proposed by GOBEL' is given as follows : For the titrations a solution of soda, of which 1000 c.c. =. 100 grams absolute acetic acid ; a solution of phosphoric acid which titrates to phenol-phthalein of a strength equal to the soda solution ; and a solution of hydrochloric acid which titrates to litmus of a strength equal to the soda solution. A weighed quantity of the acetate under assay is treated with some measured quantity taken as an excess of the standard phosphoric acid ; the mixture evapo- rated to dryness ; the residue treated with water and evaporated again, and until the odor of acetic acid is no longer obtained ; the residue then treated with water and the mixture titrated for excess of phosphoric acid, with the standard soda, using phenol- phthalein, and noting the result in equivalent of acetic acid. Subtracting this figure from that for the acetic acid represented by the phosphoric acid first added, the difference is the figure for the acetic acid in the acetate taken subject, however, to cor- rection for free lime and lime carbonate in acetate of lime taken for assay. By titrating a weighed portion with the standard hydrochloric acid, adding an excess, expelling carbon dioxide, and bringing back to the neutral tint of litmus with standard soda, the acetic acid equivalent to the unsaturated earthy bases is found, and deducted for the correction. A rapid method of assay, "which has been much used, but is apt to give figures too high, is carried as follows : A weighed quantity of the acetate of lime is supersaturated with a known quantity of sodium carbonate in solution ; the precipitate filtered out and washed ; and the alkali of the total filtrate estimated as sodium carbonate by titration of an aliquot part. The loss of sodium carbonate due to the removal of acetic acid (and acid empyreuma) in the precipitation is calculated into acetic acid, and figured upon the quantity of acetate of lime taken. BLAIR (1885, where cited) obviates the difficulty of the color of the 1 1884 : Repert. '/. anal. Chem., 3, 374 ; Zeitsch. anal Chem., 23, 264. H ACETIC ACID. solution by filtering it through animal charcoal, and then obtains good results by this method. g. Commercial Grades and Common Impurities. The strengths of acetic acid have been designated by a " No.," alto- gether different from vinegar numbers, but probably originating, under the British excise system, in the number of parts of four per cent, vinegar producible by dilution. 1 Thus No. 8 acid is that which diluted to eight parts will have about four per cent, strength. The two grades numbered on this system, in this coun- try, are " No. 8 " and " No. 12." Interpreted according to original intent, therefore, No. 8 should be of 32 per cent., and No. 12 of 48 per cent, strength. Dr. Squibb finds that the best qualities of No. 8 acid actually prove of near 30$ strength, bearing label mark of s.g. 1.040 ; the poorer qualities of No. 8 are near 25$ strength, and issued without a gravity mark. No. 12 acid is less common, and often runs from 38 to 40 per cent, of real acid. The strengths of vinegar numbers refer, in the British custom, to the number of grains of dried sodium carbonate neutralized by one Imperial fiuid-ounce. pTagCOg : C 2 H 4 O 2 : : 53 : 60 : : 1 : 1.132. The number x 1.132 = grains absolute acid per fluid-ounce (of grains 437.5 X s -g-) The number X 0.259 = grams absolute acid in 100 c.c. vine- gar. In this country vinegar numbers have been grains of sodium bicarbonate neutralized by one fluid ounce, wine mea- sure. NaHCO 3 : C 2 H 4 O 2 : : 84 : 60 : : 1 : 0.7143. The number x 0.7143 = grains absolute acid per fluid-ounce (of grains 455.7 X s.g.) The number X 0.1567 '= grams absolute acid in 100 c.c. vinegar. Much of the " Glacial Acetic Acid " of commerce is not over 75 per cent, of real acid (SQUIBB). It can easily be furnished of 99-)- per cent., as required by U. S. Ph. Of impurities in ordinary acetic acid, the more common are mineral acids, especially hydrochloric, empyreumatic bodies, and metallic salts. Empyreuma, and other foreign bodies having odor or taste, are recognized by these senses after neutralizing with potassa or soda. " When diluted with five volumes of distilled water, the color caused by the addition of a few drops of test-solution of permanganate of potassium should not be sensibly changed by standing five minutes at the ordinary tem- perature (absence of empyreumatic substances)." U. S. Ph. According to Dr. Squibb, when 1 c.c. of the acid, diluted with 5 1 SQUIBB, 1883 : Ephemeris, i, 258. 2 1883 : Ephemeris, I, 260. VINEGAR. 15 c.c. distilled water, is treated with 3 drops of decinormal solu- tion of permanganate, in comparison with the same addition to the distilled water, if the color does " not become fully brown " within ten minutes, it is " a very good acid indeed," but the glacial acid " should stand this modification of the permanga- nate test for more than an hour." In vinegar the most common impurities are (1) free mineral acids, and (2) empyreumatic bodies (in " wood vinegar "). Be- sides, various made-up vinegars, and forms of diluted acetic acid, are substituted for or added to cider-vinegar. The absence of free mineral acid is shown by an alkaline re- action of the ash. Let the residue be carefully ignited and the cold ash touched with wet litmus-paper. The residue can be ignited on the loop of platinum wire. All natural vinegars con- tain some alkali acetate, and in absence of mineral acids will give an alkaline reaction in the ash. If tle vinegar be a mere diluted acetic acid, as a " white vinegar," a few drops of decinormal solution of fixed alkali are to be added before the evaporation, when a neutral reaction of the ash indicates free mineral acid. To estimate the quantity of free mineral acid, take 50 grams of the vinegar, add of decinormal alkali from the burette enough to surely neutralize all free mineral acid, still leaving the reaction acidulous, evaporate, ignite with care against loss, and titrate back with decinormal acid. Then c.c. -^ alkali c.c. -^ acid X 2x0. 0049 per cent, of free mineral acid, as sulphuric acid. IJsing the factor 0.00364, the statement is obtained for hydro- chloric acid, etc. Free sulphuric acid, in absence of chlorides, may be separated and determined as follows : 100 c.c. are evapo- rated on the water-bath nearly to dryness, treated with about 100 c.c. of alcohol, the mixture filtered, the alcohol evaporated off, and the residue diluted for the gravimetric estimation of the sulphuric acid in it, by precipitation with barium chloride. If chlorides be present in the vinegar, it is necessary to add silver sulphate before adding the alcohol, when both the free sulphuric and hydrochloric acids of the vinegar are estimated as sulphuric acid. It must be remembered that sulphates and chlorides are liable to be present in legitimate vinegars, and the simple reactions with silver and barium, as prescribed for acetic acid, are not applicable in tests of vinegars in general. But, according to DAVENPORT/ " in a pure cider vinegar, nitrate of silver, nitrate 1 " Report of Inspector of Vinegar of the City of Boston," 1884, p. 4 ; of Inspector of Milk of the same, 1885, p. 10. 1 6 ACETIC ACID. of barium, or oxalate of ammonium added after an excess of am- monia water, will neither of them give more than the slightest perceptible reaction." Also, " a drop of it in a loop of platinum wire, when ignited in a Bunsen lamp-flame, gives a pure potash flame without any yellow soda rays visible." "The addition of any practical amount of a commercial acetic acid to tone up the strength will give another color to the flame." Cider-vinegars yield a residue " always soft, viscid, mucilaginous, of apple flavor, somewhat acid and astringent to the taste." " If any corn glucose is present, the residue, when ignited in the platinum loop, will emit the characteristic odor of burning corn ; and if the glucose was manufactured with the commercial sulphuric acid derived from copper-pyrites, it will, as the last spark glows through the carbonized mass, emit the familiar garlic odor of arsenic." The percentage of solids in cider-vinegar, by weight of residue, is generally required to be as much as 1.5 per cent. Dr. DAVENPORT (1885) recommends that the legal limit be 2 per cent. " When 20 grams of the vinegar are mixed with 0.5 c.c. of barium nitrate test-solution (1 to 19) and 1 c.c. decinorrnal silver nitrate solution, the filtrate from the mixture should give no re- action for chlorine or sulphuric acid. When two volumes are added to one volume of sulphuric acid and then one volume of ferrous sulphate solution poured over, no brown zone should appear between the layers. The evaporation-residue from 100 grams should riot exceed 1. 5 grams. The residue should not have a sharp taste, and its ash should have an alkaline reaction." Ph. Germ. The required strength of vinegars is given by U. S. Ph. of 1870 at 4.6$; Br. Ph., 5.41#; Ph. Germ., 6$; the "proof vinegar" of British Excise, 6$, or English "No. 24." In exe- cution of the British law against adulterations of foods, the minimum limit of strength has been held at 3$. For " cider- vinegar," the limit recommended by Dr. Davenport, in the Bos- ton City inspection, is 5 per cent, of real acetic acid ; and the lowest limit there proposed, 4 per cent. In New York City the legal requirement, well enforced (1886), is 4^ per cent, of acetic acid as a minimum for all vinegars, and 2 per cent, of solids for cider- vinegars. The following is the form of Inspector's Record and Analyst's Report, under the regulations of the city of Boston, 1886 : " Vinegar : Date, ; Time, ; Proprietor's name, - ; No. , Street ; Sold by ; Price paid, - ; Quan- tity, pint ; Wholesaler's name, ; Price paid ditto, ; ACONITE ALKALOIDS. 17 District, ; Cider, ; White wine, - ; How marked. Analysis : Analysis No. - ; Acetic acid, - ; Residue, ; Character of residue, ; Chlorine, - ; Sulphates, ; Calcium^ ; Color, ; Free acid." ACIDS OF THE FATTY SERIES, CnH 2 nO 2 . See FATS. ACONITE ALKALOIDS. Natural alkaloids of plants of the genus Aconite (Ranunculacese), and artificial products of these alkaloids represented bj Aconitine, C 33 H 43 ]TO 13 = 645 (WEIGHT, 1877). CONTENTS : Chemical constitution ; saponification changes ; list of alka- loids with rational formulae ; dehydration changes ; list of alkaloids with phy- siological effects; sources; yield. Analytical outline for crystallizable and for amorphous alkaloids of aconite : a, heat-reactions of each ; b, taste and phy- siological effects ; c, solubilities ; d, qualitative tests, with limits ; e, separa- tion in general, from aconite root, from animal tissues; /, quantitative methods, gravimetric, volumetric, of produced benzoic acid ; g, commercial grades and values. Chemical constitution and character. It has been established by "Wright and his co-workers * that the crystallizable alkaloids of the aconite group are salts, or esters, of benzoic acid (or a derivative of this acid), and are readily saponifiable ~by action of alkalies or strong acids, to some extent even by water with heat. And the saponitication results in the removal of either benzoic acid or a derivative of benzoic acid, and the formation of amorphous alkaloids in place of the crystallizable alkaloids sapo- nified. The tendency of aconite alkaloids to become amorphous, with diminished physiological activity, is explained by saponifica- tion. x Their liability to another and less obvious class of chemi- cal changes, leaving them still crystallizable and with little loss of physiological activity, is shown by the proof that, l>y action of strong acids, they suffer dehydration and form apo- alkaloids. That is to say, alkalies, with more or less readiness, and even hot digestion with water, cause saponification; and strong mineral acids, even concentrated organic acids in a degree, cause both saponification and dehydration to apo-compounds. 2 Various 1 C. R. A.WRIGHT, in part with A. P. LUFF, and with A. E. MENKE, 1877- 1879 : Jour. Chem. Soc., 31, 143 ; 33, 151, 318 ; 35, 387, 399. Phar. Jour. Trans. [3] 8, 164-167. Further, MANDELIN, 1885 : Archiv d. Phar. [3] 26, 97, 129, 161 ; Phar. Jour. Trans. [3] 15. JUERGENS, 1885 : Phar. Zeitsch. Russland. 2 It is a noteworthy correspondence that three active alkaloidal agencies of intense physiological power, in eytensive medicinal use at present, Aconitine, i8 ACONITE ALKALOIDS. other transformations are brought about by agents not so com- monly employed in processes of separation as are the alkalies and acids. The following equations show the changes of saponifioation, by alkalies or acids, upon four of the crystallizable alkaloids of the aconites according to Wright : 1 Crystallizable alkaloids. Amorphous alkaloids. Benzoic acid. C 33 H 43 NO 12 (acomtine)+H 2 O=C 26 H 39 NO n (aconine)+C 7 H 6 O 2 (picraconitine)-]-H 2 O =C 24 H 41 NO 9 (picraconine-f-C 7 H 6 O 2 1 (japaconitine)-f-3H 2 O = 2C 26 H 41 NO 10 (japaconine)+2C 7 H 6 O 2 (pseudaconitine)-}- H 2 O Dimethylprotocatechuic acid. = C 27 H 41 ]SrO 9 (pseudaconine )+ C 9 H 10 O 4 The rational formulae a of these alkaloids include the an- hydride of benzoic acid, or of one of its derivatives, in the crystallizable members of the group ; and include hydroxyl instead of the acid anhydride in the amorphous members of the group (WEIGHT) ; as follows : Aconitine, C 33 H 43 NO 12 =C 26 H 35 NO 7 (OH) 3 . . (C 7 H 5 O) Aconine, C 26 H 39 NO n = C 26 H 35 1S T O 7 (OH) 3 , OH Japaconitine, C^H^O,-, =2 [C 06 H 39 NO 7 O . O . (C 7 H 5 O)] Japaconine, C 96 H 41 NO; o =C 26 H 39 ~lS T O 7 O . (OH) 2 Pseudaconitine, 3 C 36 H 49 NO 12 =C 27 H 37 NO 5 (OH) 3 . . (C 9 H 9 O 3 ) Pseudaconine, C 27 il 41 lS[O 9 =C 27 H 37 ]NO 5 (OH) 3 . OH. Atropine, and Cocaine, agree in being saponifiable alkaloids easily giving up either benzoic acid or some near derivative of benzoic acid. (Atropine : KRAUT, 1865. Cocaine : LOSSEN, 1865. Aconitine : WRIGHT, 1877.) Among other saponifiable alkaloids, yielding acids of the aromatic group, are piperine, and certain veratrum alkaloids. 1 In saponiiication by alkali, the benzoin acid or its derivative is left in com- bination with the alkali, from which it is obtained by acidulation. In saponi- fication by acid the amorphous alkaloid is obtained in salt of the acid. 2 WRIGHT (1879), in his last contribution upon the aconite alkaloids, strongly inferred the existence of a "hypothetical parent-base, C33H 47 NOi2" =C 2 6H 39 N07.(OH)3.0.(C 7 H50). JUERGENS (1885, where before quoted), by a modified process of extraction from the root, and thorough purification, ob- tained aconitine which, in elementary analysis, gave him numbers for C33H 47 NO]2. The alkaloid gave the intense numbing sensation upon the tongue, without a recognizable bitter taste. 3 MANDELIN (1885), by investigations (without elementary analysis), con- cluded that aconine and pseudaconine are the same, so that, in his view, aconi- tine and pseudaconitine differ only by their acidulous radicals as found by Wright. ACONITE ALKALOIDS. The amorphous alkaloids are found in the plant, as well as obtained by alteration of the crystallizable alkaloids during sepa- ration from the plant. The changes of dehydration to apo-alkaloids, by action of acids, is shown by the following comparisons of rational for- mulae : Aconitine, C 33 H 43 NO 12 =C 06 H 35 :NO 7 (OH) 3 . . (C 7 H 5 O) Apo-aconitine, C 33 H 41 ]S r O 11 ^C 26 H 35 NO 7 (OH)O . O . (C 7 H 5 O) Aconine, C 26 H 39 NO 11 =C 26 H 35 ^O 7 (OH) 3 . OH Apo-aconine, C 26 H 37 NO 10 =C 26 H 35 NO 7 (OH) 2 O Pseudaconitine, C 3Q H: 49 ^O :Lf> ^C 27 }I 37 NO 5 (OH) 3 . 0. (C 9 H 9 O 3 ) Apo pseudaconitine, ^30^4^0^ =C 27 H 37 N0 6 (OH)0 . . (C 9 H 9 3 ) The natural alkaloid, japaconitine, has the constitution of a sesqui-apo-derivative. Chief Sources of the Natural Aconite Alkaloids. A. Naellus* root. u Aco- Aconitine. Aconine. Pseudaconitine ) in small propor- Pseudaconine \ tiori, if at all. nite" of U. S. Ph. and Ph. Germ. A. ferox, root. " Indian Aconite " " Nepal Aco- nite." Bish, or Bikh. " Himalaya root." Japanese aconite, root. A. lycoctonum? root. A. anthora, root. A. paniculatum, root. Pseudaconitine. Pseudaconine. Japaconitine. Japaconine. Other alkaloids. Aconitine. Pseudaconitine. Amorph. alkaloids. Pseudaconitine, Amorph. alkaloids. Picraconitine. 1 A report of alkaloids from this getic effect like curare DRAGENDORFF plant, amorphous, and having an ener- & SPOHN, 1884. 2O ACONITE ALKALOIDS. The chief Aconite Alkaloids : Synonyms, Crystallization, and Activity. ' Name. Synonyms. Formula. Crystallization. Physiolog. effect. Aconitine. Cryst. aconitine. Napaconitine. C 33 H 43 N0 12 Crystallizable, when free, as well as in salts. Of typical aco- nite activi- ty. Pseudaconitine. Napelline. t^eraconitine. Acraconitine. English aeon. C 36 H 49 N0 12 Base and its salts crys- tallize with difficulty. Approaches to or equals the activity of aconitine. Japaconitine. Cryst. alkaloid of Japanese root. C e .H 88 N.O,, Crystallizable both free and in salts. Closely resem- bles aconi- tine in pro- perties and effects. Aconine. Amorphous aco- nitine. A pro- duct of aconi- tine, by alkalies or acids. C 26 H 39 NO n Amorphous, both free and in salts. ./ s Of far low- er activity than aconi- tine. Bitter. Pseudaconine. Amorphous aco- nitine. A pro duct of pseud- aconitine, by alkalies. C 27 H 41 N0 9 Amorphous, free or com bined. Of far low er activity than acorn tine. Bitter. Japaconine. Amorphous alka- loid of Jap. aconite. Pro- duct of Japaco- nitine. C 26 H 41 N0 10 Amorphous, free or com- bined. Closely resem- bles aeon in e in properties and effects. Picraconitine. Inactive, bitter al- kaloid of A. pa- niculatum and other species. C3iH 4 5N0 10 Base crvst. with diffi- culty. Salts crystallize well. Bitter. Not poisonous. Picraconine. Amorphous pro- duct of picraeo- nitine. C 24 H 41 N0 9 Amorphous. Bitter. Not poisonous. Apo-aconitine. Product of aconi- tine, by action of acids. C 33 H 41 N0 11 Crystallizable. Of the same activity as aconitine. Corresponding apo-derivatives, by action of acids on Pseudaconitine, Aconine, etc. (See p. 19.) ACONITE ALKALOIDS. 21 For medicinal uses the II. S. Ph. and Ph. Germ, admit only the tuberous root of A. Napellus ; the Br. Ph., also U. S. Ph. of 1870, admit both " root " and leaf of A. Napellus ; the Ph. Fran, authorizes the use of root and leaf of A. Napellus and A. ferox. It is understood that both Japanese aconite root l and root of A. ferox are largely used for the manufacture of medicinal alkaloid "aconitine." A. Sterkeanum contains poisonous alkaloids. Yield of natural Aconite Alkaloids. WRIGHT obtained, in 1876, from A. Napellus only 0.03 per cent, of pure aconitine, and only 0.07 per cent, of total alkaloids free from other matter. Again, from Japanese aconite roots 0.18 per cent, of mixed al- kaloids. JUERGENS (1885) obtained, by a modified Duquesnel's process, of thoroughly purified aconitine (for elementary analy- sis) 0.02 per cent. By chemical assays (1883) LABOKDE and DUQUESNEL found in A. Napellus root, of " crystalline alkaloids " from 0.05 to 0.40 per cent., averaging 0.15 per cent. ; of " amor- phous, insoluble substance " having an effect like aconitine in kind, "a few" tenths per cent.; and of " amorphous, soluble, bit- ter substance," about 1.5 per cent. ZINOFFSKI," working by volu- metric estimation with Mayer's solution (probably an inexact measure of total aconite alkaloids) in A. Napellus and other species, from the fresh leaf (calculated to basis of dry material) 0.73 to 1.38 per cent, total alkaloid ; from the fresh stalks, 0.25 to 0.90 per cent. ; and from the fresh flowers, 1.51, 1.65, and 5.52 (!) per cent, total alkaloids. HAGER (1863) reported fi nd- ing in the best commercial root of A. Napellus from 0.64 to 1.25 per cent, [total alkaloids]. SQUIBB (1882) found the leaf of A. Napellus to have only about one-ninth of the physiological effect of the same quantity of the root. CULLAMORE (1884) found the action of A. ferox to be more intense in degree than that of an equal quantity of A. Napellus. The " aconitine " of the market may contain any mixture of the aconite alkaloids frequently aconitine, japaconitine, pseud- aconitine, and the wholly amorphous alkaloids. Systematic phy- siological assay of four commercial grades of " aconitine," by Dr. SQUIBB in 1882, in comparison with good powdered aconite root, gave the following results : (1) Of unknown make had only the physiological potency of the root ; (2) " Ordinary," 8 times the strength of the same weight of the root ; (3) Pseudaconitine, 83 times the power of the root ; (4) " Crystallized," 111 times the 1 Respecting Japanese and Chinese Aconites, see LANGGARD, also WASO- vicz, 1880 : Archiv d. Phar., 14, 217, and 15, 161 ; Phar. Jour. Trans., [3] 10, 149, 1020 ; Proc. Am. Pharm., 29, 170-182. 2 Dragendorff's " Werlhbestimmung," 1874, p. 13. 22 ACONITE ALKALOIDS. effect of the root. If we accept Wright's analyses, first above given, the total aconite alkaloids should have from 500 to 1400 times the potency of the same weight of root. Further, Dr. Squibb found that the article (4) was a nitrate containing not more than 80.7 per cent, of hydrated alkaloid. Aconitine was dropped in the last revision of the U. S. Ph. and in the last re- vision of the Ph. Germ. It is retained by Br. Ph. and Ph. Fran. THE CRYSTALLIZABLE ACONITE ALKALOIDS are identified by their organoleptic effect (b), the agreement of their precipitations (d 9 p. 25) and solubilities (c), and by yielding benzoic acid or its derivative when saponified (p. 18, and under j). THE AMORPHOUS ALKALOIDS of the aconites are distinguished from the crystalliz- able ones by greater solubilities in water (/r., 36, 535. FERDINAND F. MAYER, 1862-3: Pro. Am. Pharm. 1862, 238; and Chem. News, 7, 159; 8, 177, 189 ; Am. Jour. Phar., 35, 20 Zeitsch. anal. Chem, 2, 225; Jahr. Chem., 1863, 703. G. DRAGENDORFF. "Werthbestimmung," 1874, p. 9 and elsewhere. A. B. PEESCOTT, "Estima- tion of alkaloids by potassium mercuric iodide," J880: Am. Chem. Jour., 2, 294; Jour. Chem. Soc., 42, 664: Chem. News, 45, \U;Ber. d. Chem. Oea., 14, 1421. 44 ALKALOIDS. solution proposed by Mayer is the one generally used for pur- poses qualitative or quantitative. It is a deciiiormal solution of (HgCl 3 +6KI) = the hydrogen equivalent of Hg. Of dry, crystallized mercuric chloride, 13.525 grams ; potassium iodide, 49.680 grams ; separately dissolved in water, and the mixed solu- tions made up to one liter. The reactions of the solution appear to correspond with the formula KIHgI 2 .(KI) 3 +2KCl, 1 instead of (KI) 2 HgI 2 +2KI+2KCl. Dragendorff prefers to make quan- tities, above specified, to 2 liters instead of 1. Mayer's solution is applied only in acidulous solutions, in testing for alkaloids ; therefore ammonia does not interfere, as the precipitate of mercurammonium iodide is not formed in pre- sence of free acids. The acidulation may be with sulphuric or hydrochloric acid, and may be strong without dissolving the precipitate. The solution " tested must not be alcoholic, and must not contain acetic acid. Some organic matters other than alkaloids cause precipitates. With strychnine the precipitate is obtained in dilution of 1 to 150000 ; with quinine, in solutions of about the same dilution ; while with morphine, or with atro- pine, solutions of 1 to 4000 do not give the precipitate. The precipitates are curdy or flocculent, and for the most part of a yellowish-white color. Caffeine and theobromine are not precipitated by potassium mercuric iodide. The composition of some of the alkaloid iodomercurates varies with conditions of concentration, excess of reagent, and acidity ; while the precipitates of other alkaloids are nearly con- stant in composition. With strychnine the precipitate is not far from C 21 H 22 N 2 O 2 HIHgI 2 a ; with morphine the precipitate cor- responds to a variable mixture of (C 17 HpNO 3 ) 4 (HI) 4 (HgI 2 ) 3 and (C 17 H 19 NO 3 ) 4 (HI) 6 (HgI 2 ) 3 ; with quinine the precipitation ap- pears to be most nearly, though not closely, represented by (C 20 H 24 N 2 O 2 ) 2 (HI) 3 (Hgl 2 ) 3 ; and with atropine the gravimetric value of the precipitate does not correspond to its volumetric factor. In volumetric use the " end-reaction " is denoted only by the 'In the proportions for (KI) 2 HgI 2 +2KCl, the mercuric iodide remains dis- solved only in concentrated or hot solution. The quantity of alkali iodide adopted by Mayer cannot be very much reduced and retain solubility at the decinormal dilution in the cold. A permanent solution with the help of bro- mide can be obtained as follows: HgCl 2 + 4KI + KBr : mercuric chloride, 13.525 ; potassium iodide, 33.12 ; potassium bromide, 5.94; water to 1000 by volume. This solution may be supposed to contain (KI) 2 HgI 2 + KBr + 2KCL (The author, 1880: Am. Ghem. Jour., 2, 304.) 2 The author, 1880 : Am. Chem. Jour., 2, 296. GENERAL REAGENTS. 45 completed precipitation. ' After the last addition from the bu- rette the precipitate is either allowed to subside, or a little por- tion is filtered out and a drop of the reagent added from the bu- rette to the clear solution. Some of the precipitates subside readily, strong acidulation usually favoring this result ; with others much time is required, and titration in this way is gene- rally slow. Filtration is the better way : using a minute niter, not over 5 millimeters or J inch in radius, held in a loop of pla- tinum wire or a coil of drawn-out glass tubing, over a glass slide placed upon black paper. A drop or two is taken, with the stir- ring-rod, from the mixture containing the precipitate, filtered through the wet filter, and treated over the black ground with a drop of the reagent from the burette, when the slightest turbid- ity can be seen. Before the end of the titration all the test-por- tions are drained and rinsed with a few drops of water passed through the filter into the mixture containing the precipitate. In volumetric estimation the strength of the alkaloidal solu- tion should usually be 1 of alkaloid to 200 of solution a second estimation being made, if need be, for this graduation. The quantity of alkaloid precipitated by 1 c.c., under given condi- tions of concentration, etc., is stated with the directions for quan- titative work on the several alkaloids described in this work. A quite full list of the volumetric factors for Mayer's solution was given by Mayer, and some of these have been subjected to con- trol analyses by Dragendorff and others ; but the presentation of such a list is here intentionally avoided. It must be understood that the alkaloidal equivalent of one c.c. varies with the condi- tions, especially with that of concentration. Unless the analyst has good authority for an alkaloid equivalent, given with speci- fied conditions, he should standardize his Mayer's solution, with an alkaloid solution of known strength, for himself, holding de- grees of concentration, acidulation, mass, and time the same for the titration of the solution of unknown strength that they are for the solution of known strength of alkaloid. The end of the reaction is the point when further addition of the reagent ceases to cause a precipitate. Before this point is reached, however, in some cases the addition of a drop of the solution of the alkaloid will cause a precipitate the mixture having attained a composi- tion of equilibrium (not very rare among chemical reactions) in which precipitation is caused by a drop of either the iodomercu- rate or the alkaloid solution. When the precautions here re- 1 Trials of various indicators for the end-reaction were reported by the author in Am. Chem. Jour. , 2, 304, where also attention is called to the error of Mayer's direction to titrate back with silver nitrate. 46 ALKALOIDS. quired are observed, titration with Mayer's solution becomes a trustworthy means of estimation. The alkaloids can be obtained from their iodomercurate pre- cipitates by triturating the washed precipitate with stannous chloride solution and potassium hydroxide to strong alkaline reaction, and then exhausting with ether or chloroform or ben- zene as a solvent for the alkaloids. Strong alcohol can be used as a solvent if potassium carbonate be taken instead of potassium hydroxide. Also, the mercury can be removed from the precipi- tates by dissolving in alcohol, adding acid if need be, treating with hydrogen sulphide gas, and filtering. The filtrate can be freed from iodine, if this be desired, after expelling the hydrogen sulphide, by adding some excess of silver nitrate solution, filter- ing, adding hydrochloric acid to the filtrate, and filtering again. (3) Phosphomolybdate * A fixed alkali phosphomolybdate in strong nitric acid solution in effect a solution of phosphomo- lybdic acid. Applicable in acidulous solutions and in absence of ammo- nium salts and free ammonia, which also precipitate it. It is prepared as follows : The yellow precipitate formed on mixing acid solutions of ammonium molybdate and sodium com- mon phosphate the ammonium phosphomolybdate is well washed, suspended in water, and heated with sodium carbonate until completely dissolved. The solution is evaporated to dry- ness, and the residue gently ignited till all ammonia is expelled, sodium being substituted for ammonium. If blackening occurs, from reduction of molybdenum, the residue is moistened with nitric acid and heated again. It is then dissolved with water and nitric acid to strong acidulation ; the solution being made ten parts to one of the residue. It must be kept from contact with vapor of ammonia, both during preparation and while pre- served for use. The precipitates of alkaloids, by adding this reagent to their acidified solutions, are amorphous, and of yellowish colors, some- times orange-yellow, in other cases brown-yellow. In general they have very little solubility, and are obtained in very dilute solutions. Besides ammonia, other bodies not alkaloids are liable to give precipitates with this reagent. A negative result is trust- worthy for the exclusion of more than traces of alkaloids in the solution tested. Most of the precipitates are soluble in ammonia, and those of alkaloids that are strong reducing agents mostly dis- 1 SONNENSCHEIN, 1857: Ann. Chem. Phar., 104, 45. DE VRIJ: Jour, de Pharm., 26, 219. STRUVE, 1873: Zeitsch. anal. Chem., 12, 170. GENERAL REAGENTS. 47 solve with the blue color of reduced molybdic acid, or with some shade caused by admixture of blue. The ammoniacal solution is blue with aconitine, aniline, atropine, berberine, morphine, nicotine, and physostigmine. Alcohol and ether do not dissolve the precipitates, and acetic acid has but a slight solvent action. The alkaloids can be recovered from the precipitates by add- ing potassium or sodium hydroxide solution, and shaking out with an immiscible solvent for the alkaloid, as ether, chloroform, benzene, or amyl alcohol. Adding potassium carbonate instead of hydroxide, strong alcohol can be added instead of an immis- cible solvent. A gravimetric value of the phosphomolybdate precipitate has been obtained for a few of the alkaloids, but it has not been ascertained what conditions are necessary to secure a constant composition. 1 (4) Bromine in aqueous hydrobromic acid. WOEMLEY di- rects the use of aqueous hydrobromic acid saturated with bromine. Applicable to aqueous solutions of the salts of the alkaloids, neutral or slightly acidulous witli a mineral acid, and in absence of acetic acid and of alcohol, which dissolve the precipitates. Besides alkaloids, the phenols and other bodies give precipitates with bromine. (See Phenol.) The limit of precipitation of the alkaloids is at dilution to from 5000 to 100000 parts with mor- phine, 1 to 2500 ; with nicotine or conine, 1 to 10000 ; with aco- nitiiie, codeine, or brucine, 1 to 25000 ; with strychnine, narco- tine, or veratrine, 1 to 100000 (WOEMLEY). In general the pre- cipitates are amorphous ; with atropine, crystalline. (5) Potassium cadmium iodide (MAEME, 1866). Prepared by saturating a boiling concentrated solution of potassium iodide with cadmium iodide, and adding an equal volume of cold-satu- rated solution of potassium iodide. In diluted solution, precipi- tation is apt to occur. This reagent precipitates the aqueous so- lutions of alkaloid salts, acidified by sulphuric acid, the precipi- tates being soluble in excess of the precipitant, or in alcohol. Amorphous at first, the precipitates become crystalline. The al- kaloids can be recovered from the precipitates as directed for those formed by potassium mercuric iodide. (6) Potassium bismuth iodide (DEAGENDOEFF, 1866). Pre- pared from bismuth iodide, in the way directed for the last- 1 It appears probable that a dilute solution of the phosphomolybdatp. standardized by solution of an alkaloid of known strength, could be used 1o estimate the quantity of the same alkaloid under strictly parallel conditions. The end-reaction can be found as directed for Maver's solution. 48 ALKALOIDS. named reagent. Cannot be diluted. Applicable as a precipitant to aqueous solutions of alkaloid salts, strongly acidified with sulphuric acid. (7) Metatungstic acid, Phosphotungstic acid (SCHEIBLER, I860), Silicotungstic acid (GODEFFROY, 1876), and Phospho-anti- wwmc acid (SCHULTZE, 1859), have been used as general preci- pitaiits for the alkaloids. GODEFFROY (1877) uses a solution of ferric chloride in hydrochloric acid as a precipitant for alkaloids. (8) Tannic acid (BERZELIUS, HENRY, DUBLANC, HAGER), in solution with 8 parts of water and 1 part of alcohol, gives whitish, grayish-white, or yellowish precipitates with nearly all the alka- loids. In the larger number of instances these precipitates are easily soluble in acids, frequently dissolving in excess of the tannic acid ; on the contrary, some of the alkaloids are precipitated by tannic acid only in strong acid solutions. Ammonia dissolves the tannates of the alkaloids. Dilute acetic acid dissolves the precipitates of tannates of aconitine, brucine, caffeine, colchicine, morphine, physostigmine, and veratrine ; acetic acid not dilute, the precipitate of quinine. Cold dilute hydrochloric acid does not dissolve the precipitates of tannates of aconitine, berberine, brucine (dissolves sparingly), caffeine, cinchonine, colchicine (dissolves slightly), narcotine, papaverine, thebaine, solanine, strychnine (dissolves slightly), veratrine. Cold dilute sulphuric acid does not dissolve the pre- cipitates of tannates of aconitine, physostigmine, quinine, sola- nine, veratrine. Precipitates are completely formed in solutions strongly acidulated with sulphuric acid, by aconitine, physostig- mine, and veratrine, though none of these alkaloids gives a full precipitate in slightly acidulated solution. Alkaloids are recovered from their tannates by mixing the moist precipitate with lead oxide or carbonate, drying the mixture, and extracting with an immiscible solvent or with alcohol. (9) Picric acid, HC 6 H 2 (]SrO 2 ) 3 O (WORMLEY, 1869 ; HAGER, 1869). Used in very dilute, saturated aqueous solution, or in a sparing addition of the alcoholic solution. Applied as a preci- pitant of alkaloids in their neutral solutions, or, better, in solu- tions acidulated with sulphuric acid. Many of the precipitates become crystalline, and give characteristic forms under the mi- croscope ; in general they have a yellow or yellowish- white color. "With morphine the precipitate is formed, in drop-tests, in solu- tion of 1 to 500 ; with aconitine, atropine, or veratrine, in solu- tion of 1 to 5000 ; with brucine or narcotine, in a solution of GENERAL REAGENTS. 49 1 to 20000 ; with strychnine, 1 to 25000 ; with nicotine, in solution of 1 to 4000 (WORMLEY). The alkaloids can be re- covered from their picrate precipitates by adding an alkali solu- tion and exhausting with a solvent immiscible with water, or by evaporating to dry ness with a solution of potassium or sodium carbonate, and extracting with alcohol. HAGER has used preci- pitation with picrate in some estimations of alkaloids. For cinchona alkaloids, 1 10 grams of the powdered bark, covered with 130 c.c. water, with 20 drops of caustic potassa solution of s.g. 1.3, are digested at boiling temperature and stirred for a quarter of an hour. Of dilute sulphuric acid, s.g. 1.115, 15 grams are added, and the mixture boiled 15 to 20 minutes. "When cold the whole is made up, by the addition of water, to 110 c.c. (" the volume of 110 grams of water "). The mixture is filtered, through a paper filter of 10.5 to 11.0 centimeters (8J inches) diameter, into a graduated jar, and the volume of the filtrate (about 60 c.c.) noted. To this filtrate (100 c.c. of which represents the 10 grams of bark) picric acid solution saturated in the cold is added, in quantity about 50 c.c., or enough to complete the precipitation (as ascertained by allowing a few drops to flow down the side of the vessel). After half an hour the precipitate is gathered on a weighed filter, washed, and dried between blotting-papers over the water-bath. The dried preci- pitate of picrates of cinchona alkaloids contains (according to Hager) two molecules of picric acid as anhydride, 440 parts, to one molecule of cinchona alkaloid, 308 to 324 parts, without water of crystallization. Or 8.24 parts of the precipitate indi- cate about 3.5 of mixed cinchona alkaloids. Then the noted number of c.c. of decoction taken : 100 : : the indicated quan- tity of mixed alkaloids in the precipitate : x = quantity mixed alkaloids in the 10 grams of bark. (10) Platinic Chloride. Auric Chloride. Solutions of these salts, hardly to be classed as special reagents for alkaloids, yet give precipitates with the greater number of them. Platinic chloride is often required in establishing distinctions between alkaloids, as noted in this work under the qualitative reactions of the respective compounds. The same may be said of auric chloride. The melting points of the alkaloidal compounds of these metals serve as constants useful for identification, especially in distinguishing the derivatives of alkaloidal radicals. The composition of these metallic precipitates has in most cases been '1869 : Phar. Centralh., p. 145 ; Zeitsch. anal. Chem., 8, 477. 50 ALKALOIDS. estimated from the percentages, respectively, of metallic platinum and metallic gold, left after ignition. These percentages were much depended upon in the earlier years of the chemistry of the alkaloids, and are given full and prominent statements in Gme- lin's Hand-book of Chemistry. The platinum precipitates are divisible into those which do and those which do not dissolve in hydrochloric acid cinchonine and quinine, morphine, and strychnine being placed among those not readily soluble in this acid. The platinum precipitates have a yellow or yellowish color. The gold precipitates of a number of the alkaloids blacken by reduction on standing. Color-reactions of the Alkaloids. In general it should be borne in mind that color-reactions are subject to variation (1) by impurities of the alkaloidal material, (2) by impurities of the reagent, and (3) by conditions of concentration, mass preponde- rance, temperature, and time. Also, that the best authority to guide the operator is the result of a control-test upon a known portion of the alkaloid in question, holding all conditions to be the same. Concentrated Sulphuric Acid 1 dropped upon the dry alka- loid, on a white porcelain surface or on glass over a white ground, without heating, reacts as follows : colorless with atro- pine, caffeine, chelidoiiine, cinchonidine, cinchonine, codeine, hyoscine, hyoscyamine, morphine, nicotine, pilocarpine, quini- dine, quinine, staphisagrine, strychnine, theobromine. Of these, on warming, a purplish to brow r n color is given by morphine. Yellowish colors are given by colchicine, gnoscopine, and jer- vine ; reddish colors are given (either at once or after a short time) by apornorphine, brticine (pale rose), conine (pale), gelse- minine, mecoiiidine, narceine (to black), narcotine (yellow-red to violet and blue), nepaline, physostigmine, rhoeadine, sabadilline, sabatrine, solanine, taxine, thebaine, veratrine, and veratroidine ; bluish colors are given by cryptopine, curarine (on standing), and papaverine ; and greenish colors by beberine, berberine, emetine (brown to green), piperine, pseudomorphine, and rhoeadine. Of glucosides, reddish colors (mostly bright) are given by amygda- lin, % colombin, cubebin, elaterin, hesperidin, phloridzin, populin, salicin, sarsaparillin, senagin, smilacin, syringin, tannic acids. 1 Traces of nitric acid, not infrequent as an impurity in " C. P. sulphuric acid," cause a great difference in the reaction with morphine and other alka- loids colored by nitric acid. See the composition of " Erdmann's reagent," given in the foot-note under Nitric Acid Color Tests. On the Reactions of Alkaloids with Sulphuric Acid, cold, warm, and hot alone, with nitric acid, and with permanganate see GUY, 1861-2: Phar. Jour. Trans, [2], 2, 558, 662 ; 3, 11, 112 ; Zeitsch. anal. Chem., i, 90. GENERAL REAGENTS. 51 Froehdds Reagent concentrated sulphuric acid containing molybdio acid. 1 A solution of 0.001 gram of molybdic acid or alkali molybdate, in 1 c.c. of concentrated sulphuric acid (DRA- GENDORFF), freshly prepared by the aid of heat, and used when cold. FKOEHDE took 0.005 gram of the molybdate to 1 c.c. of sulphuric acid, and BUCKINGHAM took as much as 1 part of molybdate to 15 of the sulphuric acid ; but the more attenuated proportion of the molybdate (1 to 1840) gives the more dis tinctive reactions. The reduction of molybdic acid to hydrated molybdic molybdate is attended with a bright blue color. This reduction occurs in concentrated sulphuric acid, by heat alone, at the temperature of incipient vaporization of the sulphuric acid. Numerous inorganic and organic reducing agents cause the reduction and give the color to molybdate. As a character- izing reaction it is applied mostly to alkaloids, when iion-alka- loidal matter must be excluded, and the more dilute solution of molybdate is the more trustworthy. Froehde's reagent gives no color with atropine, caffeine, cin- chonidine, cinchonine, conine, delphinine, hyoscine, hyoscya- mine, nicotine, strychnine, theobromine ; yellowish colors with aconitine, colchicine, piperine ; reddish colors with brucine, eme- tine (red changing to green), narceine (changing to blue), saba- dilline (reddish-violet), solanine, thebaine (orange), veratrine (gra- dually, cherry-red) ; bluish colors with codeine (gradually, deep blue), morphine (violet to blue), narceine (yellow T -brown to red and blue), staphysagrine (violet-brown) ; greenish colors with apomorphine (green to violet), beberine (brown-green), berberine (brown-green), emetine (red changing to green and turned blue by hydrochloric acid), quinine (pale), quinidine (pale). Of glu- cosides, colocynthin gives slowly a cherry-red color ; elaterin, yellow ; phloridzin, slowly, blue ; populin, violet ; salicin, violet to cherry-red ; syringin, blood-red to violet-red colors. Nitric acid, of s.g. 1.40 to 1.42, applied in a drop to the dry alkaloid upon white porcelain, gives a color, frequently reddish, with numerous alkaloids. No color is obtained with atropine, caf- feine, cinchonidine, cinchonine, conine, gelseminine, quinidine, quinine, strychnine, theobromine. Yellowish colors are obtained 1 FROEHDE, 1866: Archiv der Phar., 126, 54; Zeitsch. anal. Chem., 5, 214; Pro. Am. Pharm., 15, 241. ALMEN, 1868: N. Jahr. /. Phar., 30, 87; Zeitsch. anal. Chem., 8, 77. KAUZMANN, 1869: Zeitsch. anal. Chem., 8, 105. BUCK- INGHAM, 1873: Jour. Chem. Soc., 27, 715; Am. Jour. Phar., 45, 179. DRAGEN- DOR.FF, 1872: " Beil rage zur gericht. Chem. organ. Gifte." A. B. Prescott, 1876: " Froehde's Reagent as a Test for Morphine," Am. Jour. Phar., 48, 59; Jahr. der Pharm., 1876, 502. On various reactions of the blue oxide of molyb- denum, see MASCHKE, 1873. 52 ALKALOIDS. with aconitine (yellow to brown or red, variable), codeine (orange-yellow), morphine (yellow to red), narceine, narcotine, papaperine (orange), piperine (orange), rhoeadine, sabadilline (yellow), thebaine, veratrine. Red colors are obtained by aconi- tine (red-brown, variable), apomorphine, berberine (red brown), brucine (blood-red), papaverine (orange-red), pseudomorphine (orange red), physostigmine. A Hue color is given by colchi- cine and by solanine (Dragendorff). Some glucosides give briglit colors ; ligustrin and syringin, blue tints. Sulphuric acid (concentrated), followed by a minute addi- tion of nitric acid (s.g. 1.40-1.42), or of solid potassium nitrate. 1 No color is given by atropine, caffeine, cinclionidine, cinchonine, nicotine, pilocarpine, quinidine, quinine, staphysa^rine, strych- nine, theobromine. Red colors are given by brucine, curarine, narcotine (red- violet), nepaline, physostiginine, sabadilline, the- baine, veratrine (gradually, cherry-red). A violet color is given bv morphine (under directions specified for that alkaloid). Co- deine gives a succession of colors, as also does colchicine. Sulphuric Acid and Cane /Sugar.' 2 The substance to be tested, in the dry state, is mixed with 6 to 8 parts of cane-sugar, and a few milligrams of the mixture are placed upon a drop or two of concentrated sulphuric acid, over a white ground. The gradual browning of the sugar itself is disregarded, and will be covered by the briglit colors of characteristic reactions. No colors are given by atropine, brucine, caffeine, cinchoriidine, cin- chonine, conine, nicotine, quinidine, quinine, strychnine, and theobromine. Reddish colors are given by codeine, curarine, gelseminine, morphine (purple-red, then blue- violet, dark blue- green, and lastly blackish-yellow limit 0.0001 to 0.00001 gram), nepaline (gradually), sabadilline (reddish-violet). KlluisJi color by veratrine. Various oils, and albuminoids, give bright colors with sulphuric acid and sugar. Hydrochloric Acid, concentrated, gives colors with only a few alkaloids. Reddish colors are given by physostigmine, sabadilline, and veratrine. 1 ERDMANN, 1861: Ann. Pharm. Chem.. 120; Zeitsch. anal. Chem., I, 224. Erdmaim mixed six drops of nitric acid of s.g. 1.25 with 100 c.c. of water, and added ten drops of this mixture to 20 grains of sulphuric acid. Of this " Erdmann's reagent " 8 to 20 drops were added to 1 or 2 milligrams of the solid to be tested, and the color noted after | to ^ hour. HUSEMANN, 1863 : Ann. Chem. PJiar., 128, 303 the well-known test for morphine. DRAGEN- DORFF, 1868: "Ermittelung von Giften," p. 239. 2 SCHNEIDER, 1872: Ann. Phys. Chem. Pogg., 147, 128; Zeitsch. anal. Chem., 12, 218. Respecting reactions with substances not alkaloids, SCHULTZE, Ann. Chem. P/iar., 71, 266. MICROSCOPICAL CHARACTERISTICS. 53 Other Reagents for alkaloids as a class, or for groups of alka- loids. Iodine in hydriodic acid, gold bromide, sodium gold thiosulphate, potassium gold iodide, lead tetra-chloride, and manganese perhydroxide in sulphuric acid, were reported upon by F. SELMI in 1877. Perchloric acid, FEAUDE, 1879-1880. Sodium arseniate with sulphuric acid, TATTEBSALL, 1879. Cu- pric ammonium hydrate, NADLEE, 1874. Ferric chloride and sulphuric acid, How, 1878. Fused antimonious chloride, SMITH, 1879. Nitroferricyanide of sodium, as a precipitant, HOESLEY, 1862. The Microscopical Characteristics of alkaloids, in their various combinations, receive attention to some extent in all chemical literature upon these bodies, and in the description of the several alkaloids in this work. Among the special contri- butions are the following : HELWIG, 1865 : " Das Microscop in Toxicologie." GODEFFEOY and LEDEEMANN, 1877 : on cinchona alkaloids. WOBMLEY, 1885 : " Microchemistry of Poisons," 2d ed., Philadelphia. A. PEECY SMITH, 1886 : identification of alka- loids by crystallization under the microscope, Analyst, II, 81. On Micro sublimation of Alkaloids : HELWIG, 1864: Zeitsch. anal. Chem., 3, 43; "Das Microscop in Toxicologie," 1865. GUY, 1867 : Phar. Jour. Trans., [2], 8, 718 ; 9, 10, 58, 106, 195, 370; " Forensic Medicine," London, 1875. STODDAET, 1867. ELLWOOD, 1868. BLYTH, 1878 : Jour. Chem. Soc., 33, 313. In this work, see under Caffeine. The Subliming Cell of Dr. Guy, improved by Blyth, consists essentially of a ring of glass, about J inch in thickness, or from J to f inch. This glass ring rests on an ordinary " cover-glass " a thin disc used under this name in microscopy. Another cover-glass is placed upon the ring, which is of a diameter to fit the cover- glasses, and with them make a closed cell. The ring can be made of a section of glass tubing by grinding the edges. The cell, so constituted, was heated by Dr. Guy through a brass plate on which it rested. Dr. Blyth prefers to rest the cell upon liquid metal, using mercury for tern peratures below about 100 C., and fusible metal for tempera- tures above this point. The liquid metal is contained in a porcelain capsule of about 3 inches diameter, supported on the ring of a retort-stand, and heated directly by the flame. A flask of suitable size, from which the bottom has been removed, is placed over the capsule, upon the ring of the retort-stand, and made to carry the thermometer, held in a perforated stopper and with its bulb immersed in the liquid metal by the side of the subliming cell. A minute speck of the article tested is placed 54 ALOINS. on the lower disc of the cell. Blyth's definition of a sublimate is this : " The most minute films, dots, or crystals, which can be observed by a quarter-inch power, and which are obtained by keeping the subliming cell at a definite temperature for sixty seconds." ALOINS. Varieties of a neutral crystalline principle ob- tained from the several kinds of aloes. As first described (T. & H. SMITH, 1851), it was obtained from Barbadoes aloes, and was the body now named barbaloin. There have been described : SOMMARUGA and Aloes. Yield. 1 EGGEB, 1874. Barbaloin Barbadoes 20-25 per cent. , at most, Ci T H 2 o0 7 TILDEN, 1872. Nataloin Xatal 16-25 percent., at most, Ci 6 H 18 7 Socaloin Socotrine 3 per cent, average, CisHieOr Zanzibar PLENGE, 1885. TILDEN ascribes to barbaloin the formula C 34 H 36 Oi 4 . H 2 O, and to nataloin C 35 H 28 O 1;L ; and FLUCKIGER (1871) obtained for socaloin C 34 H 38 O 15 . 5H 2 O. Aloins are identified by their color-reactions with nitric and sulphuric acids, by which, also, and by production of chrysammic acid, they are distinguished from each other (d). ALOES is found in mixtures by treatment with acids, or by extraction with amyl alcohol and treatment with various reagents (d, p. 55). Chrysammic Acid, p. 56. As to physiological effects, with reference to valuations, 5, p. 55. a. Barbaloin,) crystallized from a concentrated aqueous solution of Barbadoes aloes, appears in tufts of small yellow prisms, losing 2.69 per cent, of water by drying at 100 C. or in vacuum. Nataloin exists in a crystalline state in Natal aloes, from which it is left on treating with an equal portion of alcohol at 48 C. or under, and when recrystallized forms thin, brittle, rectangular scales with some of their angles truncated. It loses no water at 100 C. Socaloin exists in Socotrine or Zanzibar aloes in prisms of good size ; when recrystallized from methyl alcohol, tufted acicular prisms, which may be obtained 2 to 3 millimeters long. At 100 C. it loses about 12 per cent, of water. b. Aloins are without odor and have the taste of aloes. Their purgative power has been questioned, and while they have 1 Of 18 varieties of aloes, yields of from 2.2 to 31.3 per cent, were ob- tained: DRAGENDORFF, 1874: " Werthbestimmung." ALOINS. 55 had some little medicinal use as therapeutic representatives of aloes, more in Great Britain than elsewhere, yet this use has not extended, although aloin is more agreeable for administration than the aloes from which it is extracted. DRAGENDORFF states (1874: : " Werthbestimmung "), on experimental data, that (1) the resins of any variety of aloes, separated as insoluble in cold water, in doses of 0.35 gram (5 to 6 grains), prove inactive ; (2) that perfectly pure aloins, in doses of 0.3 to 0. 5 gram (5 to 7 grains), prove inactive with many persons ; and (3) that the so- called aloes-bitter, soluble in cold water and containing either amorphous aloin or oxidized products, represents the activity of the drug ; also (4) that the purgative power of an aloes is measured by the quantity of bromaloin precipitated from an aqueous solution of the drug, also by the quantity of precipi- tate by tannic acid. Dragendorff infers that aloin is converted into bodies having the purgative action of aloes. TILDEN (1876) found that all three aloins are decidedly uncertain and variable in their action, and seem to present no advantage over an equal dose of aloes, except perhaps that griping was rather less com- mon under their use. c. The aloins are soluble in water, barbaloin the most freely of the three, socaloin in about 90 parts, and nataloin very spar- ingly. Alcohol dissolves all the aloins, socaloin requiring about 30 parts, and nataloin about 60 parts (230 parts absolute alcohol). In ether aloins are but slightly soluble, though socaloin dissolves in about 380 parts. Aloin " from the different varieties of aloes " is described in Br. Ph. (1885) as " sparingly soluble in cold water, more so in cold rectified spirit, freely soluble in the hot fluids. Insoluble in ether." d. Nitric acid (s.g. near 1.40 or 1.42), applied to the dry aloin on a porcelain slab, gives a bright red color with barbaloin or nataloin, not with socaloin. The crimson red of barbaloin fades quickly ; the blood red of nataloin does not fade unless heated (HISTED, 1871 ; TILDEN, 1876). Boiling with nitric acid produces chrysammic acid, C 14 H 4 (NO 2 ) 4 O 2 (tetranitrodioxyan- thraquinone), of intense red color, from both barbaloin and nataloin, not from socaloin. Oxalic and picric acids, in addition, are obtained from barbaloin by action of boiling nitric acid (distinction from socaloin or nataloin). If nataloin be wet with concentrated sulphuric acid, and then touched by the vapor of strong nitric acid from a glass rod or by a minute fragment of potassium nitrate, a fine blue color is obtained (distinction from barbaloin or socaloin). Concentrated sulphuric acid, applied to 56 AMYGDALIN. the dry substance, and followed by a minute fragment of potas- sium dichromate (as in the fading- purple test for strychnine), causes a green or greenish-purple color, changing to greenish- yellow. Alkalies cause the decomposition of aloins. Solu- tions of aloes, too, lose their bitterness and their purgative power when made alkaline (G. MCDONALD, 1885). CHEYSAMMIC ACID (see above) crystallizes in gold-glittering needles, or in yellow fern-leaves resembling picric acid. It de- tonates on heating. It is acidulous in reaction, and of intensely bitter taste. It is insoluble in cold water, easily soluble in alco- hol and in ether. It forms colored salts with metallic lustre. Potassium chrysammate crystallizes with bright green lustre, or (from acid solutions) as bright crimson needles with a slight golden reflection. ALOES. If a grain of aloes or dry mixture be dissolved in 16 drops of strong sulphuric acid, 4 drops of nitric acid (s.g. 1.42) added, and the mixture diluted with one ounce of water, a deep orange or crimson color will be obtained. On adding ammonia the color changes to a claret. All substances containing chry- sammic acid behave nearly the same in this test, except that they turn pink on adding ammonia directly to their aqueous solutions, while the solutions of aloes do not (Cmpps and DY- MOND, 1885). If a fluid containing aloes be extracted with amyl alcohol, the residue left by evaporating this solvent will have a bitter taste, and when this residue is dissolved in water the solu- tion will give precipitates with bromine in potassium bromide solution, basic lead acetate, mercurous nitrate, and tannic acid, and will reduce gold chloride and Fehling's solution. The dry residue will give a blood-red color with potassium cyanide and hydroxide (!)KAGENDOEFF, LENZ, 1882). AMYGDALIN. Co H 27 NO n = 457 (LiEBia and WOHLER, 1837). ^H^O^OHVp^.CN (SCHIFF, 1870). A gluco- side which occurs in the bitter almonds and in numerous other plants which yield hydrocyanic acid by natural fermentation. The bitter almonds, after removal of the oil by pressure, are di- gested twice with hot 95$ alcohol, and allowed to stand for some time. The alcohol is decanted and concentrated to a syrup, from which the amygdalin is precipitated by ether. The precipitated amygdalin is washed with ether and recrystallized from boiling alcohol. Amygdalin crystallizes from alcohol in colorless scales anhy- drous or with 2H 2 O, from water in transparent prisms, becoming opaque in the air, and containing 3H 2 O. It becomes anhydrous ARBUTIN. 57 at 110-120 C. It is odorless, of a slightly bitter taste and neutral reaction, and rotates the plane of polarization to the left. It is soluble in any proportion of hot and 12 parts cold water; in 11 parts boiling and 904 parts cold alcohol (s. g. 0.819) ; in 12 parts boiling and 148 parts cold alcohol (s. g. 0.939) ; insoluble in ether. Concentrated sulphuric acid dissolves it with violet-red color, which turns black on warming. The other mineral acids decompose it. In contact with emulsin and water (10 parts amygdalin, 1 part emulsin, and 100 parts water) it is- changed into benzoic aldehyde (oil of bitter almonds), hydrocya- nic acid, and glucose, as follows : Through farther change of the hydrocyanic acid, formic acid also is formed. By boiling with dilute sulphuric acid the same reaction takes place, when formic acid is always formed. 17 parts of anhydrous amygdalin, or about 24 to 25 parts (theoretical- ly, 19 parts) of the ordinary commercial amygdalin, yield, when fermented with emulsin, one part hydrocyanic acid and 8 parts bitter- almond oil. Boiling amygdalin with aqueous alkalies or baryta changes it to ammonia and amygdalic acid (C 20 H 26 O 12 ). ANALYSIS, ELEMENTARY. See ELEMENTARY AN- ALYSIS. ANALYSIS OF PLANTS. See PLANT ANALYSIS. ANALYSIS, ORGANIC. See ORGANIC ANALYSIS. ARBUTIN. C 12 H 16 O 7 = 272. A glucoside found (about 3.5$) in the leaves of the bearberry (Arctostaphylos Uva-ursi) and in a number of other plants, especially in those belonging to the order Ericacese. It may be obtained by precipitating the decoction with lead subacetate, freeing the filtrate from lead by hydric sulphide, treating with animal charcoal, and crystallizing. Crystallizes in bunches of silky needles which have the com- position (C 12 H 16 O 7 ) 2 .H 2 O. They become anhydrous at 100 C. and melt at 170, have a bitter taste and neutral reaction. Sparingly soluble in cold water, readily soluble in hot water and in alcohol; slightly soluble in ether. Boiled with dilute sulphuric acid, or subjected to the action of emulsin or another ferment contained in the bearberry, it is converted into hydro- quinone, C 6 H 6 O 2 , and glucose. Treated with manganese diox- ide and sulphuric acid, it is oxidized to quinone, C 6 H 4 O 2 , and formic acid. It does not reduce alkaline cupric solution, and is $8 ASPARA GINBEBIRINE . not precipitated by salts of the metals. Concentrated sulphuric acid dissolves it without color. Nitric acid turns it black, gradu- ally dissolving it to a yellow solution. If an aqueous solution be rendered alkaline with ammonia and then phosphomolybdic acid added, it becomes blue [one part in 140,000 parts water gives a distinct color JUNGMANN, 1871 : Am. Jour. Phar., 43, 205]. ARICINE. See CINCHONA ALKALOIDS. ASPARAGIN. CJIgNgOg^lSS. Amido-succinamic Acid. Exists already formed in asparagus (Asparagus qfficinalis) and a great many other plants. It crystallizes from the cold-water ex- tract of asparagus upon concentration to a thin syrup, and may be purified by treatment with animal charcoal and recrystalliza- tion from hot water. The crystals are hard, brittle, transparent prisms of the tri- metric system having the composition C 4 H 8 N 2 O 3 . H 2 O. They are odorless, have a slight, disagreeable taste, are permanent in the air, and become anhydrous at 100 C., above which temperature they are decomposed. Asparagin is soluble in 58 parts cold and 4.4 parts boiling water ; in 500 parts cold and 40 parts boiling 60$ alcohol ; in 700 parts boiling 98$ alcohol ; insoluble in abso- lute alcohol, chloroform, ether, and benzene ; easily soluble in acids and aqueous alkalies. It forms weak compounds with both acids and alkalies. In contact with the accompanying extractive substances, yeast or casein, etc., it is changed by fermentation into, succinate of ammonium (sometimes with the intervening formation of aspartate of ammonium). When boiled with acids or alkalies it is resolved into aspartic acid (C 4 H T N'O 4 ) or amido- succinic acid, and ammonia. Respecting the quantitative estimation of asparagin, see the current reports of E. SCHULZE, 1881 to 1885. ATROPINE. See MIDKIATIC ALKALOIDS. BAKING POWDERS. See TAKTAKIC ACID. BEBIRINE. Biberine, Ci 8 H 21 NO 3 , dried at 100 C. In Greenhart or Bibirin bark (British Guiana), H. EODIE, 1835 ; as "JBuxine" in bark of Buxus sempivirens or Common Box, Faury, 1830, identified with bebirine by Walz in 1860 and Fliickiger in 1869 ; as Pelosin, in Pareira Brava root (Chon- drodendron tomentosum and Cissampelos Pareira), Wiggers, 1839, identified with bebirine by Fliickiger in 1869. BEN ZOIC ACID. 59 a. A white, amorphous powder, melting at about 145 C., and decomposing at a higher temperature. Its salts of common acids are uncrystallizable, pulverulent or resinous, and white or yellowish- white. #. The alkaloid and its common salts are odorless, with a strong and persistent bitter taste. Its effect is held to resemble that of quinine, and is given in about the same quantities. c. Very slightly soluble in water (6600 parts cold, 1500 boiling) ; soluble in 5 parts absolute alcohol and 13 parts of ether ; soluble in chloroform, benzene, amyl alcohol, and carbon disul- phide. Its solutions are strongly alkaline to test-papers. The sulphate, hydrochloride, and acetate are readily soluble in water ; the solutions having a neutral reaction. d. The alkali hydrates and carbonates give precipitates, soluble in excess of the hydrates. Precipitates are caused by potassium mercuric iodide (white), potassium iodide, mercuric chloride, gold chloride (yellow- white), platinic chloride (pale yel- low), and sodium phosphomolybdate (dissolved blue by ammo- nia, decolored by boiling), picric acid (yellow), sulphocyanate (reddish-white). Nitric acid dilute and potassium nitrate give a white precipitate (Fliickiger); sodium phosphate, a white precipitate. The pure alkaloid does not reduce iodic acid. e. Bebirine has been prepared from the different plants in which it occurs, by extraction with acidulated water, and precipi- tation with soda or ammonia, with a precipitate by lead subacetate and extraction therefrom by dilute sulphuric acid (or by digesting the precipitate with magnesia and extracting with alcohol or ether). Purification by animal charcoal is sometimes used instead of, or after, the lead precipitation ; the object in either operation being chiefly removal of resinous matter. f. The precipitated alkaloid loses 8.2 p. c. water (near- ly IfHgO) at 100 C. Bebirine platinic chloride (C 18 H 21 NO 3 ) 2 (HCl) Q PtCl 4 (Bddeker). The Hydrochloride is C 18 H 21 NO 3 . HC1. The' Sulphate (C 18 H 21 NO 3 ) 2 H 2 SO 4 [Maclagen]. BENZOIC ACID. Benzoesaure. Acide Benzoique. CLH 6 O 2 =:122 (monobasic). C 6 H 5 .CO 2 H. Carboxyl-benzene. Without isomers. Sources: Benzoic acid is found, uncom- bined, in the proportion of 10 to 19 per cent., in Benzoin, the balsamic resin of Styrax Benzoin, produced in Siam and Suma- tra ; also in smaller proportions in Balsam of Peru, in Balsam of 60 BEN ZOIC ACID. Tolu ? (Bussy, 1876), in fruit of Yaccinium vitis-idaea (cowberry) (O. Low, 1879), and in the Xanthorrhcea resins. It has been found in certain plums and other fruits. In combination with ethereal bases, forming essential oils, it is found in numerous bal- sams and resins, and in the oils of cinnamon, bergamot, origa- num, and cananga (ylang-ylang). The fragrant oil slightly per- vading the Benzoin is reported to be ethyl benzoate. The benzoates frequently accompany or substitute the compounds of cinnamic acid, and sometimes occur with coumarin. The suint of sheep's wool contains benzoates (TAYLOK, 1876). Benzoic acid is slowly formed by the atmospheric oxidation of oil of bitter almond (benzoic aldehyde), appears among the oxidation-products of cinnamic acid and various aromatic compounds, and results from certain decompositions of albuminoids. SCHULZE (1885) finds benzoic acid in the heavier (phenol -containing) coal-tar oils. Hippuric acid, in decomposing urine, may change to benzoic acid. Benzoic acid is manufactured (1) from Benzoin, either, as "flowers of benzoin," by direct sublimation, 1 or in the wet way, as "crystallized benzoic acid," by dissolving with lime, precipi- tating from the calcium benzoate solution by adding hydrochloric acid, and recrystallizing from hot water to remove resin. (2) From the Hippuric acid of graminivorous animals, chiefly horses and cows, by concentrating the urine, acidulating with hydro- chloric acid to obtain crystallized hippuric acid, and boiling the latter with crude hydrochloric acid, when benzoic acid and the by-product glycocoll are promptly formed : CH 2 . KE(CO . C 6 H 5 ) . CO 2 H+H 9 O =C 6 H 5 C0 2 H+CH 2 . NH 3 . C0 2 H (3) From the coal-tar product, Naphthalene, C 10 H 8 , which by treatment with nitric acid is converted into phthalic acid, C 6 H 4 (CO 2 H) 2 , when the latter, heated to about 350 C. with its equivalent of calcium hydrate, in absence of air, forms the lime salt of benzoic acid : 2C 6 H 4 (CO 2 ) 2 Ca+Ca(OH) 2 =(C 6 H 5 C0 2 ) 2 Ca+2CaCO 3 . And (4) from Toluene, of the coal-tar distillates, C 6 H 5 .OH 3 , known as toluol, by formation of trichloro-toluenes (C 6 H 5 . CC1 3 ), and conversion of the latter to benzoic acid. The pharmaco- poeias require the " natural benzoic acid." Of " artificial benzoic 1 LOEWE, 1869, and Eump, 1878, maintain that part of the benzoic acid obtained is not ready formed in the benzoin, but requires to be separated from some combination, or union with another acid. The combination with cinna- mic acid, 3C 7 H 6 O a . C 9 H 8 2 , has been reported. BENZOIC ACID. 61 acid " the production from toluol is increasing, and little is made from phthalic acid. (See, further, under Impurities.) BENZOIC ACID may be identified by its behavior in sublima- tion (#), toward solvents and precipitants (c), in reduction to bit- ter-almond oil, and in its reaction with ferric salts (d). From Cinnamic acid it is distinguished by not being oxidized to bitter- almond oil (g) ; from Salicylic acid by the color of the ferric salt. It may be separated (e) by distillation of the free acid (2) or from its salts (1) ; by solution in solvents not miscible with water (3) ; by precipitation as free acid from aqueous solution of its salts (4) ; by sublimation (5) ; from cinnarnic acid (T) ; from milk (8) (p. 65). It can be estimated by acidimetry, and by weight of the free acid, or its lead salt (f). Directions for ex- amination are given (g) as to impurities, accompaniments, and required quality for specific uses, and with regard to the sources of its production. a. Benzoic acid appears in pearly, lustrous, friable, and flexi- ble plates or needles, or in flocculent masses of plate -like or nee- dle-form structure, of hexagonal outline. From dilute alcohol six-sided prisms are obtained. The pure acid is colorless or white ; that sublimed from benzoin is frequently yellowish to yellowish-brown, and this coloration is requisite in the descrip- tion of the German pharmacopoeia. The coloration deepens in long keeping. (See g.) The pure acid is permanent in the air. Specific gravity, 1.292, at mean temperature compared with water at 4 C. (SCHRCEDEK, 1880). It melts at 121 C. (CAR- NELLY, 1878), and (by same authority) boils at 249 C. (480.2 F.), subliming unchanged. But at 100 C., either dry or with steam, it vaporizes perceptibly, and its vapor irritates the throat and ex- cites coughing. By direct heat, alone, as in a test-tube moved over a flame, it vaporizes without residue, the sublimate, if slowly deposited, crystallizing in needles. The vapors redden litmus paper. From benzoates heated with phosphoric acid, or bisulphate, the same vapors and sublimate may be obtained. Benzoic acid is carried over, to some extent, with vapor of alco- hol, benzene, and other solvents of low boiling points. Boiled with strong alkalies in aqueous solution it suffers change. ~b. Benzoic acid has a sharp, acid taste, and when pure is without odor. The pharmacopoeia! acid, from benzoin, has an agreeable aromatic odor, slight in the acid by precipitation, strong in the acid by sublimation, sometimes resembling vanilla, and by authority of the Ph. Germ, somewhat empyreumatic. 62 BEN ZOIC ACID. That from toluol often has an almond odor ; that from hippuric acid a urinous odor. The medicinal dose of benzoic acid does not overgo 20 grains. Locally it sometimes causes mucous irri- tation. -In the human body benzoic acid is converted into hip- Suric acid, the reaction being the reverse of that given above 3. 60), and excreted in the urine. If large quantities of ben- zoic acid be administered, a portion may be carried into the urine without change. Benzoic acid is an efficient antiseptic and antiferment, more powerful than salicylic acid. ARCHER (1878) used, for infusions, saccharine liquids, etc., about 4 grains to one pound, or near 0.06 per cent. ECCLES (1885) estimates about 0.04 per cent, to be sufficient for hypodermic medicated liquids. c. Benzoic acid dissolves in water as follows (BOURGOIN, 18T9) : at 15 C. (59 F.) in 408 parts ; at 20 C. (68 F.) in 345 parts ; in 17 parts of boiling water. In 500 parts water of ordi- nary temperature (Fluckiger* s Phar. Chem.) In 372 parts water (Phar. German.) In 333 parts at 15 C. ; 250 parts at 20 C. (Hager's Commentar., 2d ed.) In 2J to 3 parts of al- cohol of ninety per cent. ; in 2.2 parts absolute alcohol ; in 1 part of boiling alcohol. In 2 to 3 parts of ether ; 7 to 8 parts of chloroform ; 8 parts of benzene. Freely in petroleum benzin, amyl alcohol, and dissolves in volatile oils and in fixed oils. Benzoic acid has a decided acid reaction to test-papers, and causes effervescence in aqueous solutions of carbonates. Carbon dioxide decomposes alkali benzoate in alcoholic solution, causing a precipitate of alkali carbonate. The metallic benzoates are normal salts of a good degree of stability. Ferric benzoate be- comes in part basic in water, and mercurous benzoate in. hot water forms mercury and mercuric benzoate. Both the normal and basic lead salts are obtained. The normal benzoates are either freely or moderately soluble in water ; those of lead, sil- ver, and mercury being sparingly soluble in hot water, but pre- cipitated by adding solutions of alkali benzoate to the metallic salt solutions in the cold. Alcohol dissolves most benzoates sparingly or freely ; it decomposes the benzoates of mercury. BENZOATE OF SODIUM crystallizes, with one aq., in slightly efflo- rescent needles ; from a drop of alcoholic solution, in microsco- pic star-form groups. The salt dissolves, with a neutral reac- tion, in about 2 parts cold water, and in 13 parts of 90$ alcohol, not in ether or chloroform. Ammonium benzoate crystallizes anhydrous ; it loses ammonia and acquires free acid when ex- posed to the air. Calcium benzoate crystallizes in feathery nee- BEN ZOIC ACID. 63 dies, with four molecules of water, efflorescent, and soluble in 20 parts of cold water. Cinchonidine benzoate, normal, forms short prisms, anhydrous, soluble in 340 parts of water at 10 C. Benzoates of methyl and ethyl are colorless, oily liquids, sinking in water, of pleasant and balsamic odors, boiling respectively at 199 and 212 C., not more than slightly soluble in water, freely soluble in alcohol. d. Aqueous solutions of benzoates, by addition of hydro- chloric acid or sulphuric acid, give a voluminous, crystalline, white precipitate of benzoic acid, subject to its solubilities as stated above (c). Ferric chloride solution, in a neutral benzoate solution, gives a flesh-colored, voluminous precipitate of basic ferric benzoate, formed more quickly if the reagent be slightly basic. The precipitate is not readily dissolved by acetic acid. Free benzoic acid, in excess of saturated solution, is slowly pre- cipitated by the normal iron salt. If the solution tested be strongly alkaline in reaction, a misleading brown precipitate of ferric hydrate may occur. The ferric succinate precipitate is red-brown. Silver nitrate, in neutral solution of a benzoate, forms a voluminous white precipitate of silver benzoate, soluble in hot water, then crystallizing on cooling, somewhat more solu- ble in alcohol, dissolved by acetic acid, also by ammonia, not ob- tained with free benzoic acid. Acetate of Lead, in neutral solution of a benzoate, not too dilute, gives a white precipitate of lead benzoate, somewhat soluble in excess of the reagent, soluble in hot water, dissolved by acetate of ammonium, not by ammonia. Treatment with hydric sulphide resolves the pre- cipitate into lead sulphide and free benzoic acid, the latter being separated by hot filtration or by help of alcohoL Also, if the lead benzoate be boiled with a requisite quantity of sodium sul- phate, transposition of the metals is effected, and a filtrate of so- dium^ benzoate may be obtained. Barium chloride and calcium chloride give precipitates only in concentrated solutions of alkali benzoates, but the precipitation is promoted by free addition of alcohol. Metallic magnesium, or aluminium, or sodium -amalgam, in solution of benzoic acid or benzoate, acidulated with only enough sulphuric acid to cause a moderate evolution of hydrogen, on standing from half an hour to several hours, effects the reduc- tion to benzoic aldehyde (C 6 H 5 . COH), bitter-almond oil, recog- nized by its odor. This distinctive reduction is also obtained by passing the dry vapors of benzoic acid through faintly ignited zinc-dust. Heated, with two or three parts of lime or with so- 64 BEN ZOIC ACID. dium or potassium hydrate, in a small distilling apparatus, a dis- tillate of benzene is obtained: C 6 H 5 CO 2 H=G\TH 6 +COo. With concentrated sulphuric acid, pure benzoic is not colored, but is dissolved. If glucose be present a blood-red color is obtained, as noted under Salicylic Acid, d. Pure benzoic acid does not discolor the permanganate solution, nor reduce the potassium cupric (Fehling's) solution when heated, nor blacken ammonia- silver nitrate. e. Separation. (1) Water cannot be evaporated from free benzoic acid without its serious waste, and it suffers a slight loss in evaporation of its solutions in alcohol, benzol, etc. For the concentration of its aqueous solution it is to be neutralized by adding just enough sodium carbonate. Ammonia is not re- tained in full combination. (2) Small quantities of free benzoic acid may be distilled over with water, and for this purpose ben- zoates may be decomposed by adding enough sulphuric acid. (3) Free benzoic acid may be obtained from any aqueous liquid by shaking with chloroform, or benzol, or ether, or carbon di- sulphide. The separation is by no means complete by one appli- cation of the solvent, and the more concentrated the aqueous solution the better. The chloroform or ether is caused to evapo- rate from the benzoic acid spontaneously or by a current of air from a bellows. Ether does not give as dry a residue as chloro- form. If the chloroform or ether or benzol solution be shaken with repeated portions of very dilute aqueous alkali, the benzoic acid is brought back into watery solution of benzoate. Also, ether, chloroform, etc., may be used upon dry materials, in sepa- rations of benzoic acid. (4) Precipitation, in a concentrated aqueous solution, by hydrochloric acid, collecting the precipitate after standing and at the coolest practicable temperature, is a convenient method of separation. The mother-liquid, or filtrate, may be shaken with chloroform to recover the acid remaining in aqueous solution. Materials such as benzoin resin may be di- gested with some excess of lime or alkali, and the filtrate of aqueous benzoate precipitated with acid, as in the manufacture of natural benzoic acid in the wet way. (5) The finely divided ma- terial may be heated, dry, for sublimation. In preparing the sublimed medicinal acid, the vapors are made to rise from a wide dish, through a porous paper diaphragm, and are collected upon the inner surface of a cone of sized paper, the edges being fitted or pasted close. The temperature of the sand-bath, or iron plate, should be kept some time at about 145 C. (293 F.), and gradu- ally raised at the close to 200 C. (392 F.), the operation requir- BENZOIC ACID. 65 ing from one to four hours. A second sublimate may be ob- tained after pulverizing the fused material and taking a fresh diaphragm. The Ph. Fran, directs the addition of an equal weight of sand to the powdered benzoin. An- analytic sublima- tion, for separation from fixed impurities may be conducted in a pair of clamped watch-glasses with ground edges well fitted, or closed with a narrow ring cut out of thin asbestos cloth. (6) Precipitation with lead acetate, as indicated under d, serves the demands of separation from substances not forming insoluble lead compounds. (7) From cinnamic acid by precipitation of the latter, in a cold neutralized solution, with manganous sul- phate or chloride, avoiding any excess of this reagent. Manga- nous benzoate dissolves in about 20 parts of water ; manganous cinnarnate is but slightly soluble in water. Ether or chloroform solution separates free benzoic from hippuric acid. (8) From milk, MEISSL (1882) adds lime to alkaline reaction, evaporates to one-fourth, adds gypsum, and dries on the water-bath. The dry mass, powdered, is extracted with alcohol, after acidulation with sulphuric acid. The alcoholic solution is neutralized with ba- ryta, concentrated, acidulated with sulphuric acid, and extracted with ether, from which the benzoic acid crystallizes almost pure. f. Quantitative. Free benzoic acid, in absence of other acids, whether taken in distillates, or residues of separative sol- vents, or in original materials, can be quite closely estimated volumetrically with a standard solution of alkali (BOOKMAN : " Untersuchungsmethoden," 1884), using litmus as the indicator. The weighed material for estimation is treated directly with an excess of the volumetric alkali measured from the burette, stirred to bring all the benzoic acid into solution as benzoate, when the liquid is titrated back with the proper volumetric acid. Each c.c. of normal solution of alkali (after deducting c.c. of normal solution of acid) 0.122 gram of benzoic acid. Taking 1.22 gram of the material, each c.c. of decinormal solution of alkali (after deducting for the acid used in titrating back)=l per cent, of benzoic acid. Benzoic acid may be weighed, directly, as C 7 H 6 O 2 . For this purpose the best form is that of good crystals, either from a so- lution or by slow sublimation. The residue obtained by spon- taneous evaporation, of chloroform, ether, or other separative solvent of free benzoic acid also a clean precipitate may be weighed. The acid is to be dried over sulphuric acid, any excess of liquid or adhering moisture being first taken up with blotting- paper. 66 BEN ZOIC ACID. Salts of benzole acid are usually treated to obtain the free acid, as above described (e), but they may be precipitated, in a neutral solution, by lead acetate, as stated under d. The lead benzoate, Pb(C 7 H 5 O 2 ) 2 , is washed with cold alcohol acidulated with one- half per cent, of acetic acid, and dried at 100 C. The weight multiplied by 0.5416 gives the quantity of benzoic acid. g m Impurities. Chemically pure benzoic acid is precisely the same in all properties, whether manufactured from the bal- samic benzoin or from urine, toluol, or naphthalene ; but a chemically pure acid has not been manufactured, on a commer- cial scale, from any source. The chief uses of benzoic acid are (1) in medicine and (2) in the production of dyes. It is used, also, for the manufacture of food flavors and as an antiseptic. For medicinal purposes the pharmacopoeias designate its source as follows : Ph. Germ. " From benzoin by sublimation . . . yellowish to yellowish-brown . . . with odor of benzoin, somewhat empy- reumatic." Br. Ph. " From benzoin ... by sublimation. Not chemi- cally pure. Nearly colorless." Ph. Fran. " From benzoin " prepared by alternative direc- tions (1) by sublimation, (2) by humid method. U. S. Ph. White scales or needles, " having a slight aroma- tic odor of benzoin." There may be two reasons for requiring medicinal benzoic acid to be sublimed from " the gum " : (1) the essential oil of benzoin obtained with the sublimed acid has a stimulant effect and an agreeable odor ; (2) by outlawing the artificial product the injurious impurities frequently present in it may be avoided. The artificial acid, quoted as " German benzoic acid," has been for several years priced at from one-third to two-thirds the value of the natural acid, quoted as "English benzoic acid." Un- doubtedly chemically pure benzoic acid will be made from hip- puric acid or from toluol (DYMOND, 1883 ; JACOBSEN, 1881), and furnished at prices lower tnan those for the natural acid. But hitherto, in any production of the artificial acid for medicinal uses, with little encouragement for open statement, there has been more effort to counterfeit the chemical impurities of the natural sublimed acid than to avoid the chemical impurities of the arti- ficial product. A chemically pure benzoic acid, from any source, is acceptable for the preparation of medicinal benzoates. In sensible properties the acid recently sublimed from ben- BEN ZOIC ACID. 67 zoin has a white or pearl color if sublimed slowly, at tempera- ture of about 125-140 C., with rejection of the last fraction of sublimate, even this, from some varieties of benzoin, being nearly colorless. But a sharp heat, of about 200 C., gives a yellowish sublimate, becoming yellowish-brown in its last por- tions, and in proportion to increase of color is the distinctness of empyreumatic odor obtained, in addition to the proper ethereal and vanilla-like odor of the benzoin obtained with colorless sub- limates. The acid sublimed from Sumatra or Penang benzoin has only a faint odor, not vanilla-like. Any empyreumatic oil pervading the crystals darkens gradually by action of air, and colorless samples of sublimed benzoic acid are liable to acquire a yellowish tint on long keeping. Benzoic acid well prepared in the wet way (p. 60) is in water-white crystals, larger and not so much in flocculent masses as the " flowers of benzoin." It has but a slight ethereal odor of benzoin, without empyreuma. But if it has not been crystallized from the precipitate it will contain much resin of benzoin, with some color, and will not dissolve clear in hot water. Artificial benzoic acid i's frequently obtained in distinct prismatic crystals of considerable size. That from hippuric acid is apt to have a horse-stable odor ; that from to- luol, an odor of bitter-almond oil ; and imitated " flowers of benzoin " may have ethereal or empyreumatic odors. Cinnamic acid is occasionally present in all varieties of ben- zoin. In sublimation it requires a higher heat than benzoic acid, and its vapors are denser. Sublimed benzoic acid with empyreu- matic odor and yellowish-brown color is likely to contain cinna- mic acid, if it were present in the benzoin. Benzoic acid from benzoin by the wet way is by no means likely to be free from cinnamic acid, if this were present in the benzoin. The impurities incidental to sources may be enumerated as follows : In natural benzoic acid by sublimation : Ethereal oil containing more or less styrol (cinnamene, C 8 H 8 ), vanillin (C 8 H 8 O 3 ) if prepared from the true Siamese benzoin (JANNASCH and RUMP, 1878), and sometimes empyreumatic distillate. Also cinnamic acid. In natural benzoic acid by the wet way : Cin- namic acid, resins, calcium chloride, ethereal oil. In the product from hippuric acid : Ammonia or nitrogenous bodies readily yielding it, substances giving the odor of urine or of the perspi- ration of the horse, hydrocyanic acid (a product of hippuric acid by heat), and chlorides. In toluol-benzoic acid: Chloro- toluenes, oil of bitter almond (benzoic aldehyde) which is formed from dichloro- toluene, while benzoic acid results from trichloro- toluene ammonium compounds, chlorides and sulphates. 68 BEN ZOIC ACID. Imitated natural benzole acid is prepared by subliming from a mixture of (odorless) artificial benzoic acid, and either benzoin or the resinous residue after sublimation of the natural acid. Also, by addition of ethereal oils, etc. Tests. For cinnamic acid, by its oxidation, giving benzoic aldehyde, with odor of bitter-almond oil. One gram of the acid (itself free from almond odor) with half as much permanganate of potassium, rubbed in a mortar with a few drops of water (LT. S. Ph.) A mixture of the acid with equal quantity of the permanganate and ten parts of water is warmed for a short time in a test-tube (Ph. Germ.) The test is delicate and sufficient, but the decoloration of a permanganate solution has ho meaning in the quest for cinnamic acid. For the ethereal and empyreu- matic oils peculiar to natural benzoic acid by sublimation (chemi- cal impurities in evidence of medicinal genuineness), their reac- tions as reducing agents upon permanganate, or upon silver in alkaline solution, are resorted to, as follows : Of the saturated water solution, when cold, 10 c.c. are treated with about 10 drops of solution of potassium permanganate (1 to 1000). With the true sublimed acid the color changes to red-brown and brown in from 1 to 2 minutes ; with natural benzoic acid by precipitation and crystallization the color changes in 4 to 8 minutes ; with various samples of artificial acid treated to imitate the natural sublimate, over 2 minutes (Hager's Commentar, 2d ed., 59). ' Boil 0.1 gram of the acid with 3 c.c. of water of ammonia ; add about 5 drops of silver nitrate solution, and then drops of diluted hydrochloric acid until a permanent and decided turbidity is just reached (while there is still a very slight excess of ammonia). With true sublimed benzoic acid the slight precipitate is not white, but yellowish. Concentrated sulphuric acid, with a smaller quantity of the benzoic acid, gives a yellowish color with the sublimed acid, becoming brown at 150 C. ; while at this high temperature the chemically pure acid remains colorless, and traces of hippuric acid give a brown to black color. For am- monium or other nitrogenous compounds accompanying an acid made from the urine, dissolve in a wide test-tube with a little alcohol and fixed alkali to strong alkaline reaction, heating to near boiling, and testing the vapor with moistened red litmus-pa- per and by the odor, for ammonia. For chlorides and sulphates, 1 Hager severely criticises the Ph. Germ, direction to give 8 hours for this reaction. Upon this and other tests of genuineness of natural benzoic acid, see LENKEN, 1882; SCHAER, 1882; SCHNEIDER, 1882; SCHICKUM, 1882; SCHACHT, 1881; JACOBSEN, 1881 ; DYMOND, 1883. CINNAMIC ACID. 69 test the saturated aqueous solution with silver nitrate solution, and barium chloride solution. For chloro-toluenes, slowly heat a portion under solid potassium or sodium hydrate (free from chloride) on platinum foil, dissolve the mass in water, filter if necessary, acidulate with nitric acid, and test with silver nitrate solution. Or apply the- blow-pipe test for chlorine, with the copper bead, as directed by the U. S. Ph. For hippuric acid, and gross organic and inorganic adulterations, heat a portion to vaporization and combustion, on platinum foil or clean porcelain. It should vaporize and burn, with only a residual stain : a coaly mass or incombustible residue indicating gross impurity. Also, apply any of the solvents of benzoic acid, chloroform, ether, benzol, or carbon disulphide. Hippuric acid is but slightly solu- ble in ether or chloroform. For hydrocyanic acid, distil a por- tion with a little water, and test the distillate for conversion into sulphocyanate. If a benzoate be tested, acidulate with sulphuric acid before distilling. The medicinal benzoates (see c. Chim., [2], 34, 339; Jour. Chem. Soc., Abs., 444. DE CONINCK, 1882-83. KNORR and ANTRTCK (positions in the structure of quino- line), 1884: Ber. deut. chem. Ges.. 17. 2870, 2032; Jour. Chem. .Soe., 1885, Abs., 273: 1884, Abs., 1378. GLAUS and others. Diquinolines : WILLIAMS, 1881, Chem. News, 43, 145: CLAUS, 1881-82; DEWAR, 1881; TRESSIDER, 1884; FISCH- ER (and Loo), 1 884, 1885 ; OESTERMAYER, 1885. KRAKAU, 1885. BEREND, HARTZ, KAHN, SPADY, EINHORN, 1885-86. MICHAEL, 1885: Am. Chem. Jour., 7, 182. "Ladenburg's Handworterbuch der Chemie," i. 243-298, ii. 532-595 (63 pages on quinoline). Summaries of progress, 1882-85: Am. Chem. Jour., 4, 64, 157; 5, 60, 72; 7, 200, (182). COMPARATIVE CHARACTERISTICS. 99 o . O 22 ? s s s t g > p 73 . S CO 3 ^ E m 0-^2 u % m * U is (M tn * ^ tn 4) ca t t-T " 121 ^ -3.2 H-* r^ ti ^^ O t/i W * r^ 3^^ e.= - M >L IS& S S ^s Ivj "S-S ^i . l!! I'Sg III 8 ill! 1 1|| tSl 111! . OD" $ <& S -*-* ^ i> cl' r Q > ^ d-2-cJ . Am. Pharm., 27, 492. The pub- lications above cited are illustrated with cuts. On the Identification of Alka- loids in general by Crystallization under the Microscope, a full report is made by A. PERCY SMITH, 1886: Analyst, u, 81 (illustrated). 102 CINCHONA ALKALOIDS. without variation, and the disturbing influence of evaporation must be prevented by the cover-glass at once. It is not prudent to base conclusions upon a resemblance to forms figured by other operators. Even slight differences in the purity of the reagent or in the atmospheric temperature may cause differences in the form or the rate of crystallization. The quinidine sulphocya- nate crystals are more characteristic than those of the other alka- loids, and the reaction with potassium iodide is likewise a favor- able one for microscopic recognition of quinidine. SEPARATION AND ESTIMATION OF CINCHONA ALKALOIDS. Se- paration of the total Alkaloids from Cinchona Barks. Cin- chona alkaloids exist in the barks in combination with the tannin known as cinchotannic acid (DE YRLJ, 1878). Kinic (quinic) i acid is also present in the bark, and, under action of certain solvents, unites with a part of the alkaloids. The cinchotan nates of the alkaloids are almost insoluble, while the kinates are solu- ble, in cold water. Acidulated water readily dissolves the en- tire alkaloids. In methods of analysis, with a few exceptions, the alkaloids are liberated by lime or other alkali, and dissolved from the powdered bark, in a free state, by alcohol, ether, or other solvent of the free alkaloids. But a removal of the alkaloids as hydro- chlorides is sometimes resorted to. The most favorable opera- tions for removal of the alkaloids from the bark may be clas- sified as follows : 1. The powdered bark is macerated in a mixture of chloro- form or ether, with alcohol and ammonia, and an aliquot part of the total liquid is taken (without washing) for the analysis (PROLLIUS, 1882; DE YRIJ, 1882; Ph. Germ.) 2. The powder mixed with lime is exhausted with ether in an extraction apparatus, Tollens's or other. 3. The powder mixed with lime is exhausted by digesting with a mixture of amyl alcohol and ether (SQUIBB, 1882), or amyl alcohol and benzene (Br. Phar., 1885). 4. The powder mixed with lime is exhausted by digesting and washing with alcohol (DE VRIJ, 1873; U. S. Ph., 1880, p. 78). 5. The acidulous decoction, in a part of the filtrate taken as a fraction of the total solution, is precipitated by picric acid, and the dried precipitate weighed (HAGER, 1869 ; given in this work under Alkaloids, p. 49). The use of an extraction apparatus, best adapted to ether as a SEPARATION AND ESTIMATION. 103 solvent, is a most rigidly exact and generally satisfactory way in this as in most solvent operations upon plants. But it loads the solution with more coloring and other extraneous matters, and takes longer, than the method placed first above. An aliquot part of the liquid, taken with due precautions, gives the operator quick and trustworthy results, and for ordinary uses this plan is here given the preference. Other operators prefer percolation or hot digestion, or both. The plans above enumerated have been carried out, in many cases with separation of the alkaloids from each other, or of the quinine from the other alkaloids, by different chemists, as follows : 1. Methods on the Plan of Prollius. 1 The directions of the German Pharmacopoeia of 1882 are in effect as follows: Pre- 1 PROLLIUS, 1881: Arch. d. Phar., 209, 85, 572; Am. Jour. Phar., 54, 59; New Bern., u, 22. J. BIEL, 1882: Phar. Zeitschr. Ruxsland, 21, 250. DE VEIJ, 1882: Jour, de Phar. et de Chim.; New Item., n, 258; Am. Jour. Phar., 54, 59. KISSEL, 1882: Arch. d. Phar., 220, 120. Ph. Germ., 1882, 63. FLUCK- IGEE, 1883: Phar. Zeit. t vol. 28 ; New Item., 12, 274. A. PETTIT, 1884. Ci- tations from above-named authorities': Zeit. anal. Ghem., 22, 132; Proc. Am. Pharm., 30, 204; 31, 133, 134. Prollius proposed the ethereal solvent mixture (making it by weight of ether 88 per cent., of ammonia-water 4 per cent., of 92 to 96 per cent alcohol 8 per cent.) for assays of the ether-soluble alkaloids only, and directed a chloroform mixture for assays of the total cinchona alkaloids. But Biel, and Kissel, and De Vrij agree in the statement that Prollius's ethereal solvent removes all the alkaloids. Prollius, however, used only half as much of the solvent as is here directed, according to De Vrij. De Vrij emphasizes the required fineness of the powder. He would prefer a less aqueous solvent, made by saturating the alcohol with ammonia, and adding the ether. Biel says the time of maceration should be four hours, neither more nor less, while DeVrij found one hour enough as shown by control experiment. Kissel obtains the quantity of the pure alkaloids by subtracting from the quantity of crude alkaloid the weight of resins, wax, etc., left on a tared filter, in filtration of a solution of the crude alkaloidal residue in diluted sulphuric acid. The chloroformic solvent of Prol- lius, above referred to, consisted of 76 percent, alcohol, 20 percent, chloroform, and 4 per cent, ammonia-water. The solution was wine-red, and to decolorize it a quantity of finely powdered calcium hydrate equal to the quantity of the bark is agitated with the decanted solution, which is then filtered, and this fil- trate is weighed to obtain an aliquot part, of the entire solvent taken. The weighed filtrate is evaporated, and the dried residue weighed as total alkaloid, not purified further. Tn the use of the ethereal solvent Prollius decanted the clear solution (as in the directions above), and then supersaturated the ethereal solution with diluted sulphuric acid, when the alkaloidal salts were found in a dense aqueous layer. The ethereal layer was removed and washed, once with 2 c.c., then with'l c.c. of water, the washings being added to the alkaloid solu- tion. From the latter the alcohol is evaporated, when ammonia is added just to alkaline reaction, and the precipitate dried in a tared capsule and weighed. The ethereal solution, if not distilled, should be evaporated in a flask or beaker of some depth to avoid creeping. The purification of the crude alkaloids is a matter distinct from the removal from the bark, and may be varied at will of the operator. The separation by shaking out with chloroform (p. 33) will gene- rally be preferred to precipitation by the Ph Germ. 104 CINCHONA ALKALOIDS. pare the solvent mixture by taking together 85 parts by weight of ether (s.g. 0.724 to 0.728), 10 parts of alcohol (0.830 to 0.834), and 5 parts of ammonia-water (s.g. 0.960), making 100 parts by weight. Treat 20 grams of the powdered cinchona with 200 grams of the solvent mixture, agitating thoroughly and repeat- edly, macerate one day, and pour off 120 grams of the clear liquid. Add 30 c.c. of decinormal ' solution of hydrochloric acid, remove the ether and alcohol by distillation or evaporation, concentrating the volume to 30 c.c., and, if necessary, add more hydrochloric acid until the solution has an acid reaction. Then filter, and when cold add 3.5 c.c. of normal solution of potassa. After the alkaloids have separated add to the clear supernatent liquid enough potassa solution to complete the precipitation. Collect the whole precipitate upon a filter, and wash with small portions of water, successively poured on, until drops of the washings, when allowed to glide over the surface of a cold-satu- rated aqueous solution of quinine sulphate, no longer produce a cloudiness. After allowing the alkaloids to drain press them gently between bibulous papers, and dry them by exposure to the air until they can be perfectly removed to a glass capsule. Then dry them over sulphuric acid, and finally to a constant weight on the water-bath. Of 200 grams total liquid, 120 grams were decanted, and 3:5:: weight obtained : #= weight of mixed alkaloids in the 20 grams of bark. Then a?x5 per cent, of alkaloids in the bark. Directions in detail for precautions against error, contri- buted by De Vrij and others fur the method of Prollius, are presented as follows (observe last two foot-notes) : The bark is to be very finely powdered. If of over 4$ total alkaloids, take 10 grams, otherwise 20 grams for an assay. Place the weighed portion of the powdered bark in a glass-stoppered bottle pre- viously tared, add of the ethereal solvent (above) 20 times the weight of the powder, take the exact total weight of bottle and contents, and agitate from time to time for four hours (BiEL. One hour, DE VRIJ. One day, Ph. Germ.) If any loss of weight is found, add of the solvent to restore it, and agitate and weigh again. Decant carefully so much of the solution as can be obtained per- fectly clear (into a flask from which ether can be distilled), and by weighing the stoppered bottle find the exact weight of the decanted liquid. Distil (or evaporate) off the ether avoiding 1 The Ph. Germ, directs to add 3 c.c. of normal solution of hydrochloric acid. Fliickiger, finding the resulting volume of liquid too small for the filtra- tion, advised the 30 c.c. of decinormal solution. Also advised the concentra- tion to a definite volume of 30 c.c., not in the official directions. SEPARA TION AND ESTIMA TION. 105 its taking fire then transfer the residual liquid to a small cap- sule tared with a short glass rod (rinsing with a little of the sol- vent), and evaporate and dry the residue on the water-bath. "Weight of alkaloidal solution decanted from the bottle : weight of total solvent taken in the bottle : : weight of residue : x = quantity of crude alkaloids in the amount of bark taken. To obtain the pure alkaloids, the residue of the crude alkaloids is dissolved in diluted hydrochloric acid, the solution filtered and the filter washed, the filtrate made alkaline with sodium hydrate and repeatedly shaken out with chloroform, the chloroi'ormic solution evaporated (or distilled) in a tared dish, and the residue dried at 100 C. and weighed. De Yrij found the pure alkaloids so obtained to be 16.5$ less than the crude in the case of a red Java bark. 2. Removal of the Alkaloids from the Bark ~by use of an Extraction Apparatus. For the use of an extraction apparatus upon cinchona bark, with ether as a solvent, the following ex- cellent directions of Professor FLUCKIGER are given : 1 Of a well- selected average specimen of the bark 20 grains are very finely powdered, moistened with ammonia-water, and, after standing for an hour, mixed with 80 grams of hot water ; it is then al- lowed to cool, subsequently mixed with milk of lime (prepared by triturating 5 grams of dry caustic lime with 50 grams of water), and the mixture evaporated on a water-bath until it is uniformly converted into small, somewhat moist, crumb like par- ticles, this is then transferred to a cylindrical glass tube about 2.5 centimeters (1 inch) wide and 16 centimeters (6.4 inches) long, the tube being fitted as the percolator of an extraction apparatus. The neck of this percolator is fitted with a rest of wire cloth, on which a disk of filtering-paper is held by a loose plug of cotton. The powder is packed quite compactly, and covered, at the top, with a plug of cotton which has been used to clean away the last traces of the bark. The percolator is put in place, under a condenser, in the extraction apparatus, into the receiver of which about 100 c.c. of ether is introduced, and the extraction is conducted, in the usual manner, over a water-bath for nearly a day, and until completed as shown by testing a little of the percolate. This may be tested, in the 1 " The Cinchona Barks," Power's translation, Phila., 1884, p. 69. Other solvents have been used on cinchona with an extraction apparatus. Chloroform is used in Carles's process (1 873 : Zeitsch. anal. Chem. , 9, 497). Methylated Ether, and doubtless alcohol or Methylated Alcohol, can be well used in a form of ex- traction apparatus that would carry over the vapor with desirable rapidity. io6 CINCHONA ALKALOIDS. ethereal solution, by about an equal volume of potassium mer- curic iodide solution. 1 When the extraction is completed, 36 c.c. of decinormal solution of hydrochloric acid (3.64 grams in 1 liter) are added to the ethereal solution in the receiver, when the ether is distilled off, and enough hydrochloric acid then added to give an acid reaction. When cold the liquid is filtered, the filter washed, and 40 c.c. of decinormal solution of soda (4 grams in 1 liter) are added. The precipitate is left at rest till the liquid above it is clear. Sodium hydrate solution (preferably of spec. grav. 1.3) is then added to complete the precipitation, the precipitate is col- lected on a filter, and gradually washed with a little cold water until a few drops of the washings, when allowed to flow on the surface of a cold-saturated neutral aqueous solution of quinine sulphate, cease to produce a turbidity. The drained precipitate contained on the filter is then gently pressed between bibulous paper, and dried by exposure to the air. It may afterward be readily removed from the paper without loss, and, after tho- rough drying upon a watch-glass over sulphuric acid, is finally, dried at 100 C. and weighed. The weight of the precipitate, multiplied by 5, will give the total percentage of mixed alkaloids in the bark. 3. The use of ethereal or ~benzolated mixture ofAmyl Alcohol to dissolve the free cinchona alkaloids, which are then trans- ferred to aqueous solution of the salts of these alkaloids. A. Squibtfs Process : 5 " Take of the powdered cinchona 5 ^rams ; lime, well burnt, 1.25 grams; amyl alcohol, stronger ether, puri- fied chloroform, normal solution of oxalic acid, normal solution of soda, and water, each a sufficient quantity or double all the quantities throughout, as well as the size of the vessels, etc., if 1 The ethereal solution may be treated according to the following direc- tions of FLUCKIGEU, or by any desired method for purifying the alkaloids from resins, etc. 2 E. R. SQUIBB, 1882: EpTiemeris, i, 106; BR. PH., 1885, 111. Squibb digests first with amyl alcohol alone, then adds ether in larger volume "to facilitate percolation and evaporation." The Br. Ph. digests with a mixture of amyl alcohol with thrice its volume of benzene. Squibb takes the alkaloids out of the amylic liquid by aqueous oxalic acid; theBr. Ph., by aqueous hydrochloric acid. Squibb purifies the total free alkaloids by shaking out with chloroform in alkaline mixture. The Br. Ph. undertakes the separation of the quinine with cinchonidine by precipitation as tartrates. then precipitating the remainder of the alkaloids from the filtrate as free alkaloids, these separations serving also to purify. The method of Dr. Squibb, in his unrivalled explicitness of detail, provides with great care against inefficient treatment. The approxi- mate separation with tartrate by the Br. Ph. corresponds very nearly to Squibb's approximate division into ether-soluble and ether-insoluble alkaloids, p. 117. SEPARATION AND ESTIMATION. 107 the barks be poor, or if it be desired to divide the errors of mani- pulation. "Add to. the lime contained in a 10 c.m. = 4-inch capsule 30 c.c. of hot water, and when the lime is slaked stir the mix- ture and add the powdered cinchona, stir very thoroughly, and digest in a warm place for a few hours or over night. Then dry the mixture at a low temperature on a water-bath, rub it to powder in the capsule, and transfer it to a flask of 100 c.c. capacity and add to it 25 c.c. of amyl alcohol. Cork the flask and digest in a water-bath at a boiling temperature and with vigorous shaking for four hours. Then cool and add 60 c.c. of stronger ether, of sp. gr. 0.728, and again shake vigorously and frequently during an hour or more. Filter off the liquid through a double filter of 10 c.m. = 4 inches diameter into a flask of 150 c.c. capacity, and transfer the residue to the filter. Rinse out the flask on to the filter with a mixture of 10 volumes of amyl alcohol and 40 of stronger ether, and then percolate the residue on the filter with 15 c.c. of the same mixture added drop by drop from a pipette to the edges of the filter and surface of the residue. Return the residue to the flask from whence it came, add 30 c.c. of the amyl alcohol and ether mixture, shake vigorously for five minutes or more, and return the whole to the filter. Again percolate the residue with 15 c.c. of the menstruum applied drop by drop from a pipette as before. Then put the filter and residue aside, that it may be afterward tested in regard to the degree of ex- haustion. " Boil off the ether from the filtrate in the flask by means of a water-bath, taking great care to avoid igniting the ether vapor, and also to avoid explosive boiling, by having a long wire in the flask. When boiled down as far as practicable in the flask transfer the remainder to a tared capsule of 10 c.m. = 4 inches diameter, and continue the evaporation on a water-bath until the contents are reduced to about 6 grams. 1 Transfer this to a flask of 100 c.c. capacity, rinsing the capsule into the flask with not more than 4 c.c. of amyl alcohol. Then add 6 c.c. of water and 4 c.c. of normal solution of oxalic acid, and shake vigorously and frequently during half an hour. Pour the mix- ture while intimately mixed on to a well-wetted double filter of 12 c.m. 4} inches diameter, and filter off the watery solution from the amyl alcohol into a tared capsule of 10 c.m. 4 inches 1 If only a very rough estimate of the total alkaloids be needed, this may be obtained by continuing the evaporation of the amyl alcohol solution to a constant weight, and subtracting from the result a half of 1 per cent, of the weight of bark uiken (SQUIBB). io8 CINCHONA ALKALOIDS. diameter. Wash the filter and contents with 5 c.c. of water ap- plied drop by drop from a pipette to the edges of the filter and surface of the amyl alcohol. Then pour the amyl alcohol back into the flask over the edge of the filter and funnel, rinsing the last portion in with a few drops of water. Add 10 c.c. of water and 1 c.c. of normal solution of oxalic acid; again shake vigo- rously for a minute or two, and return the whole to the wetted filter and filter off the watery portion into the caps ale with the first portion. Return the amyl alcohol again to the flask, and repeat the washing with the same quantities of water and normal oxalic acid solution. When this lias drained through, wash the filter and contents with 5 c.c. of water applied drop by drop from a pipette. Evaporate the total filtrate in the capsule on a water- bath at a low temperature until it is reduced to about 15 grams, and return this to a flask of 100 c.c. capacity, rinsing the cap- sule into the flask with 5 c.c. of water. Add 20 c.c. of puri- fied chloroform, and then 6.1 c.c. of normal solution of soda, and shake vigorously for five minutes or more. While still inti- mately mixed by the shaking pour the mixture upon a filter 12 c.m. = 4f inches diameter, well wetted with water. When the watery solution has passed through, leaving the chloroform on the filter, wash the filter and chloroform with 5 c.c. of water applied drop by drop. Then transfer the chloroform solution, by making a pin-hole in the point of the filter, to another filter of 10 c,m. = 4 inches diameter, well wetted with chloroform, and placed over a tared flask of 1 00 c.c. capacity. Wash the watery filter through into the chloroform- wet filter with 5 c.c. of the purified chloroform, and, when this has passed through into the flask, wash the chloroform-wet filter also with 5 c.c. of chloro- form applied drop by drop to the edges of the filter. When the whole chloroform solution of alkaloids is collected in the flask t boil off the chloroform to dryness in a water-bath, when the alka- loids will be left in warty groups of radiating crystals adhering over the bottom and sides of the flask. Place the flask on its side in a drying-stove, and dry at 100 C. to a constant weight. The weight of the contents multiplied by 20 gives the percentage of the total alkaloids of the cinchona in an anhydrous condition, to within 0.1 or 0.2 of a per cent, if the process has been well managed." B. J3r. Ph. Process. (1) For Quinine and Cinchonidine : " Mix 200 grains [or 12.5 grams] of the (red) cinchona bark, in 1 For an " Estimation of the Quinine," as represented by an ether-soluble division of the alkaloids, following the above method, by the same author, see p. 117. SEPARATION AND ESTIMATION. 109 No. 60 powder, with 60 grains [or 4 grams] of hydrate of cal- cium ; slightly moisten the powder with \ oz. [14 c.c.] of water; mix the whole intimately in a small porcelain dish or mortar ; allow the mixture to stand for an hour or two, when it will present the characters of a moist, dark-brown powder, in which there should be no lumps or visible white particles. Transfer this powder to a six-ounce flask [one of about 170 c.c. capacity], add 3 fluid ounces [85 c c.] of benzolated amyl alcohol [amyl alcohol, 1 volume ; benzene of sp. gr. about 0.850, 3 vols.], boil them together for about half an hour, decant and drain off the liquid on to a filter, leaving the powder in the flask ; add more of the benzolated amyl alcohol to the powder, and boil and de- cant as before ; repeat this operation a third time ; then turn the contents of the flask on to the filter, and wash by percolation with the benzolated amyl alcohol until the bark is exhausted. If during the boiling a funnel be placed in the mouth of the flask, and another flask filled with cold water be placed in the funnel, this will form a convenient condenser which will prevent the loss of more than a small quantity of the boiling liquid. Introduce the collected filtrate, while still warm, into a stoppered glass separator ; add to it 20 minims [1.1 c.c.] of diluted hydro- chloric acid [of 10.58$ real acid] mixed with 2 fluid-drachms [7 c.c.] of water; shake them well together, and when the acid liquid has separated this may be drawn off, and the process repeated with distilled water slightly acidulated with hydrochlo- ric acid, until the whole of the alkaloids have been removed. The acid liquid thus obtained will contain the alkaloids as hy- drochlorates, with excess of hydrochloric acid. It is to be care- fully and exactly neutralized with ammonia while warm, and then concentrated 'to the bulk of 3 fluid-drachms [about 10 c.c.] If now about 15 grains [0.972 gram] of tartarated soda [potas- sium sodium normal tartrate], dissolved in twice its weight of water, be added to the neutral hydrochlorates, and the mixture stirred with a glass rod, insoluble tartrates of quinine and cin- ch onidine will separate completely in about an hour ; and these collected on a filter, washed, and dried, will contain eight-tenths of their weight of the alkaloids, quinine and cinchonidine, which [in grains] divided by 2 [or in grams multiplied by 8] repre- sents the percentage of those alkaloids. The other alkaloids will be left in the mother-liquor." (2) For total alkaloids: "To the mother-liquor from the preceding process add solution of ammonia in slight excess. Collect, wash, and dry the precipi- tate, which will contain the other alkaloids. The weight of this precipitate [in grains] divided by 2 [or, in use of the metric no CINCHONA ALKALOIDS. quantities, its weight ingrains multiplied bj 8], and added to the percentage weight of the quinine and cinchonidme, gives the percentage of total alkaloids." 4. The use of alcohol to dissolve the free cinchona alkaloids, then obtained by precipitation from aqueous solution? The di- rections of the II. S. Ph. are as follows : " For total alkaloids : Cinchona, in No. 80 powder, and fully dried at 100 C., 20 grams ; lime, 5 grains ; diluted sulphuric acid, solution of soda, alcohol, distilled water, each a sufficient quantity. Make the lime into a milk with 50 c.c. of distilled water, thoroughly mix therewith the cinchona, and dry the mixture completely at a temperature not above 80 C. (176 F.) Digest the dried mix- ture with 200 c.c. of alcohol, in a flask, near the temperature of boiling, for an hour. When cool pour the mixture upon a filter of about six inches (15 centimeters) diameter. Rinse the flask and wash the filter with 200 c.c. of alcohol, used in several portions, letting the filter drain after use of each portion. To the filtered liquid add enough diluted sulphuric acid to render the liquid acid to test-paper. Let any resulting precipitate (sulphate of calcium) subside ; then decant the liquid, in portions, upon a very small filter, and wash the residue and filter with small por- tions of alcohol. Distil or evaporate the filtrate to expel all the alcohol, cool, pass through a small filter, and wash the latter with distilled water slightly acidulated with diluted sulphuric acid, until the washings are no longer made turbid by solution of soda. [Alternative directions, from this point, given below. .] To the filtered liquid, concentrated to the volume of about 50 c.c., when nearly cool, add enough solution of soda to render it strongly alkaline. Collect the precipitate on a wetted, filter, let it drain, and wash it with small portions of distilled water (using as little ~DE VRIJ, 1873: Phar. Jour Trans , [8], 4, 241; Proc. Am. Phar., 22, 268. U. S. Ph., 1880, p. 78. A. B. Prescott in "Report on the Revision of the U. S. Ph.," New York, 1880, p. 26. This is a direct and simple method, in common use and giving good results. The precipitation and washing is open to the objection, elsewhere noted, that quinine thereby suffers a little loss. This is avoided in the alternative modifica- tion by shaking out the total alkaloids with chloroform, given here from the " Report on Revision." GCEBEL (1884: Proc. Am. Phar., 32, 474) proposes, very properly, the adap- tation of the process to the plan of taking an aliquot partot the alcoholic solu- tion, as in Prollius's method, as follows: " Place 15 grams of cinchona treated with milk of lime and perfectly dried in a flask, add 150 c.c. of alcohol, weigh the whole, digest the loosely stoppered flask and contents for about two hours at 150 to 160 F., cool, replace the slight loss of weight by alcohol, filter, through a covered filter, 100 c.c. equivalent to 10 grams of the bark, and proceed with this extraction practically as directed by the Pharmacopoeia." SEPARATION AND ESTIMATION. in as possible) until the washings give but a slight turbidity with test-solution of chloride of barium. Drain the filter by laying it upon blotting or filter papers until it is nearly dry. " Detach the precipitate carefully from the filter and transfer it to a weighed capsule, wash the filter with distilled water aci- dulated with diluted sulphuric acid, make the filtrate alka- line by solution of soda, and, if a precipitate result, wash it on a very small filter, let it drain well, and transfer it to the capsule. Dry the contents of the latter at 100 C. (212 F.) to a con- stant weight, cool it in a desiccator, and weigh. The number of grams multiplied by five (5) equals the percentage of total alkaloids in the cinchona." Alternative directions from point above noted : Concentrate the filtrate to the volume of 50 c.c. or less. Transfer, rinsing with a little water, to a glass separator of 100 to 150 c.c. ca- pacity. Add solution of soda in decided excess, then at once add 30 to 40 c.c. of chloroform, stopper, agitate for a few minutes, set aside for an hour or two, and draw off the clear chloroform layer. In the same way extract with three smaller portions of the chloroform, using in all at least 120 to 130 c.c. of this solvent. The chloroform is then recovered by distilla- tion or is slowly evaporated, the concentrated liquid is trans- ferred, with chloroform rinsing, to a weighed dish, and the re- sidue dried on the water-bath to a constant weight. The grams multiplied by 5 express the percentage of total alkaloids in the bark. 1 SEPARATION OF CINCHONA ALKALOIDS FROM EACH OTHER. I. SEPARATION OF QUININE. A. From other cinchona alkaloids in general* 1. By crystallization of the sulphate in aqueous solution (p. 113). 2. " crystallization of herapathite (under Quinine, f, "Hera- pathite " ). 3. " solution in ether (p. 116). 4. " solution in ammonia (see under Quinine, <7, "Kerner's Test"). 1 "This I find," shaking out with 50 c.c. and then with three successive portions, each of 25 c.c. of chloroform, "will bring back invariably 5.99 out of 6.00 grams of pure mixed alkaloids, and is decidedly the most accurate method, given practice in the way of shaking, etc., so as to get the chloroform to settle quickly." J. MUTER, 1880: The Analyst, London, 5, 223. 2 There may be added, for trial, (5) separation by precipitation as Oxalate Shimoyama, 1885: Archiv d. Phar., [3], 23, 209. 112 CINCHONA ALKALOIDS. B. From Cinchonidine. 1. By recrystallizations of the sulphate (p. 113). 2. " solution in ammonia, in filtrate from saturated sulphate (p. H7). C. From Quinidine. 1. By non-precipitation with potassium iodide (see Quinidine, /*). 2. " non-solution of the sulphate in chloroform. 1 3. " precipitation as normal tartrate. JX From Cinchonine. 1. By non-solution of the sulphate in chloroform. 1 2. " precipitation as neutral tartrate (compare " Separation of Cinchonidine," p. 118). E. From Amorphous Alkaloids. 1. By crystallization of the sulphate. II. SEPARATION OF CINCHONIDINE. A. From other cinchona alkaloids in general. 1. After removal of Quinine, by precipitation with normal tar- trate (p. 118). 2. By precipitation as tartrate, followed by removal from Qui- nine by I. A. 1, 2, 3, or 4. B. From Cinchonine and Quinidine. 1. By non-solution of the sulphate in chloroform. 2. " precipitation as neutral tartrate (p. 118). C. From Quinine. 1. By non-crystallization as sulphate, repeated (p. 113). 2. " solution in excess of ammonia after filtration of sulphate (p. D. From Amorphous Alkaloids. 1. By crystallization as normal tartrate (p. 119). 1 Taken separately, quinine sulphate and cinchonidme sulphate each re- quires about 1000 parts of chloroform for solution, while quinidine sulphate dissolves in 20 parts, and cinchonme sulphate in 60 parts, of this solvent. Taken in mixtures of quinine or cinchonidine with quinidine or cinchonine, these differences of solubility are seriously diminished (the author with Mr. Thum, 1878: Proc. Am. Pharm., 26, 831). SEPARATION OF QUININE. 113 III. SEPARATION OF CINCHONINE. A. From other cinchona alkaloids in general. 1. By non-solution in ether (p. 116). 2. " more sparing solution in alcohol. B. From Quinine. 1. By not crystallizing as sulphate (see below). 2. " solution of the sulphate in chloroform (see note on p. 112). C. From Cinchonidine. 1. By solution of the sulphate in chloroform. 2. " non-precipitation as neutral tartrate (p. 119). D. From Quinidine. 1. By non-precipitation with potassium iodide (Quinidine, e). E. From Amorphous Alkaloids. 1. By dilute alcohol. 2. " ether. ^ SEPARATION OF QUININE (I. A, 1) from other cinchona alka- loids in general, by crystallization of the sulphate in aqueous solution The solubilities of the sulphates of the four alkaloids in water at 15 C. (59 F.) is, respectively, quinine, 740 parts; quinidine and cinchonidine, each 100 parts ; cinchonine, 70 parts. The comparative insolubility of quinine sulphate in cold water is the most trusty factor in Kerner's test for quinine, offi- cial in U. S. Ph., in Ph. Germ, since 1872, and in the Ph. Fran., 1884. 1 Sulphate insolubility also enters into the Br. Ph. test The insolubility of quinine sulphate is not materially affected by the presence of other cinchona alkaloidal sulphates, 2 which is, unfortunately, not true of the solubility of quinine in ether, or of the insolubility of quinine sulphate in chloroform. To effect complete separations, however, several recrystalliza- tions are necessary. Cinchonidine certainly opposes some resist- ance to separation from quinine. HESSE has recently reaffirmed 3 that quinine sulphate is fully freed from as much as 2 per cent, of cinchonidine sulphate by two crystallizations from boiling water. , 1862: Zeitsch. anal. Chem., 1, 150; Phar. Jour. Trans., [2], 4, 19; Am. Jour. Phar., 34, 417. 1880: Archiv d. Phar., [3], 16, 186; 17,438; Jour. Chem. Soc., 40, 63. Kerner rests the separation in good part upon the action of ammonia in the filtrate. 2 The author and Mr. Thum, 1878: Proc. Am. Pharm., 26, 834. " Report' on Revision U. S. Ph.," 1880, pp. 29, 116. 3 HESSE, 1886: Phar. Jour. Trans., [3], 16, 818; (1885) [3], 15, 869. 114 CINCHONA ALKALOIDS. DAVIES (1885) l found numerous recrystallizations necessary to obtain' a salt with constant rotatory power. KEENER (1880) a found that three to six crystallizations of commercial quinine sul- phate suffice to give a perfectly pure salt, as shown by a constant behavior in his ammonia titration. KEENER (1880) further states that in crystallizing from hot watery solution a slightly basic salt is crystallized. In this case the cleaned crystals become slightly alkaline to test-paper, while the filtrate becomes acidu- lous to a corresponding degree. To effect the utmost separation by one crystallization it is in- dispensable to hold the reaction of the initial solution exactly neutral, as a slight acidity increases the solubility of quinine sul- phate. In separations for estimation, therefore, the reaction should be neutral But in separation to prepare absolutely pure quinine salt, though at expense of partial loss, crystallization from acidulous solution is more efficient. DE YRIJ has advised to convert to the definite acid sulphate [by adding as much more sulphuric acid as the quantity required to convert the free alka- loids into neutral salts] ; then crystallize the acid salt, recrystal- lizing as necessary ; and finally form the normal salt by precipi- tating one-half as the hydrate, and dissolving the washed precipi- tate in solution of the remaining acid salt. The following directions for separation of quinine as sul- phate are in effect those of the U. S. Ph., 1880 (p. 79), 3 but with provision for better regulation of the use of acid and alkali, an increase of temperature in the digestion before crystallizing, and the drying to anhydrous instead of effloresced sulphate. The unchanged pharmacopceial text is enclosed in quotation-marks. ' ' To the total alkaloids from 20 grams of cinchona, previously weighed," or to a weighed quantity (0.5 to 5.0 grams) of any ordinary mixture of free cinchona alkaloids, taken in a weighed beaker of capacity of about 120 fluid parts for 1 part of alka- loids, add from a burette decinormal solution of sulphuric acid until the liquid is " just distinctly acid to litmus-paper " and re- tains this degree of acidity after 15 to 30 minutes' digestion on 1 DAVIES, 1885: Phar. Jour. Trans., [3], 16, 358. To same effect, OUDE- MANS, Jahr. Chem., 1876. It is surmised that a double sulphate of cinclioni- dine and quinine crystallizes, according to KOPPESCHAAR (1885) with 6H 2 0. See, also, YUNGFLEISCH, 1887: Phar. Jour. Trans. [3] 17, 585 2 KERNER, 1880: Archiv d. Phar., [3], 16, 191. As to the ammonia test, see under Quinine, g, "Kerner's test." Kerner found that heating the solution before the crystallizing at 15 C. had little influence on the result. 3 Given first in the author's contribution to "Report on Revision U. S. Ph.," 1880, p. 26. Data taken from the report of Prescott and Thum, 1878: Proc. Am. Pharm., 26, 834. SEPARATION OF QUININE. 115 the water-bath. 1 Add now decinormal solution of soda from the burette until after stirring the reaction is " exactly neutral to the test-paper." Note the number of c.c. of acid and of alkali which have been added. 2 Add water " to make the whole weigh seventy times 3 the weight of the alkaloids." Heat to near boil- ing for five or ten minutes, " then cool to 15 C. (59 F.) and maintain at this temperature for half an hour. If crystals do not appear the total alkaloids do not contain quinine in quantity over eight per cent, of their weight (corresponding to nine per cent, of sulphate of quinine, crystallized). If crystals appear in the liquid pass the latter through a filter not larger than necessary, prepared by drying two filter-papers of two to three and a half inches (5 to 9 centimeters) diameter, trimming them to an equal weight, folding them separately, and placing one within the other so as to make a plain filter fourfold on each side. When the liquid has drained away wash the filter and contents with distilled water of a temperature of 15 C. (59 F.), added in small portions, until the entire filtered liquid weighs ninety times 4 the weight of the alkaloids taken. Dry the filter, with- out separating its folds," at 100 C., 5 " to a constant weight, cool, and weigh the inner filter and contents, taking the outer filter for a counter weight. To the weight of" anhydrous sulphate of quinine so obtained add 16.89 per cent, of its amount for water of crystallization. 6 "And add 0.12 percent, of the weight 7 of J If the alkaloids be contaminated with resin and kinic acid, add enough more of the volumetric acid to surely dissolve all the alkaloids, avoiding excess of acid, and filter through a filter as small as possible, washing with the least quantity of hot water and a few drops of acid from the burette, until a drop or two of the washings cease to react for quinine when tested with a drop of May- er's solution. To the filtrate add of decinormal solution of soda as many c.c. as have been added of the acid beyond the point of just perceptible acidity, bring- ing back the reaction to this point. 2 Then (c.c. of decinormal acid c.c. of alkali) X 0.03 [0.0324 to 0.0294]= nearly the quantity of total cinchona alkaloids present, in grams. But observe that if the alkaloids have been precipitated with soda, incomplete washing may have left behind sufficient alkali to affect the result. 3 Or to make the whole measure of c.c. a number equal to 2.1 X (c.c. of decinormal acid c.c. of alkali). 4 0r until the liquid measures, in c c., 2.7 times (c.c. of decinormal acid c.c. of alkali). 6 The U S. Ph. directs to dry at 60 C. (140 F.) to a constant weight as effloresced sulphate (2H 2 0). Of this weight 11.5 per cent, is added to give the quantity of crystallized salt (7H 2 0). 6 To represent seven molecules, or 14.45 per cent, crystallization water. See under Quinine ( f). 7 That is, add 0'.0012 of weight of crystals for each c.c. of total filtrate. This correction presumes that the saturated solution ( of the filtrate) shall carry in solution 0.135 per cent, of crystallized salt (1 to 740), and that the washings (f of the filtrate) shall hold 0.067 per cent, of cryst. salt, which is ii6 CINCHONA ALKALOIDS. the entire filtered liquid (for solubility of the crystals at 15 C ) " The sum equals the quantity of quinine as crystallized sulphate in the mixed alkaloids taken. If from 20 grams of the bark, multiply by 5 to convert to percentage. Of the crystallized sul- phate (7H 3 O), 74.31 per cent, is anhydrous quinine. Separation of Quinine (I. A, 3) from other cinchona alka- loids by ether. The ether solubilities of the alkaloids taken sepa- rately are for good ether very nearly as follows, at 15 C. : quinine in 25 parts of the ether, quinidine in 30 parts, cinchonidine in 188 parts, and cinchonine in 371 parts of ether. The amorphous alkaloids of cinchona have in general a very considerable solu- bility in ether. Quinidine occurs in so small quantities that its solubility is not regarded. But the different factors of solubility above stated are not available for separation, because, as every analyst experiences, they do not hold true in mixtures of the alkaloids. Thus in a mixture of quinine and cinchonidine, qui- nine is less soluble and cinchonidine is more soluble in ether than when these alkaloids are taken separately. 1 Nevertheless, sepa- ration by ether has been in use by quinologists more than any other separation. An analyst of bark learns by the manufac- turer's results so to adjust the application of the ether that, for example, about as much of quinine will remain undissolved as there is of cinchonidine in solution. The use of ether in testing quinine for presence of cinchonine is credited to LiEBiG 2 in the test which bears his name. For use of ether in the assay of the mixed alkaloids for quinine, or for ether-soluble alkaloids, the author prefers the very practical directions of Dr. SQUIBB, who prefaces the following instructions 3 by the statement that u it half saturation. The degree of partial saturation of the washings (if held at 15 C.) is subject to the rate of application of the wash-water and its retention in the filter. Six experiments of the author with Mr. Thum (1878: Proc. Am. Pharm., 26, 834) gave a mean result equivalent to 0.00095 of crystallized sul- phate for each c.c. of filtrate (stated as 0.00085 of effloresced sulphate for each c.c.) The exact average figures as 0.00081 of effloresced salt for each c.c. J. MUTER (1880: Analyst, 5, 224) adds 0.000817 of crystallized sulphate for each c.c. of total filtrate, a filtrate which is about 80 per cent, saturated solution and 20 per cent, washings. Further evidence on the rate of this correction is desira- ble. The 0.12 per cent, correction may be too large. It is stated by CARLES (1872) that the solubility of the quinine sulphate is diminished by presence of ammonium sulphate ; by SCHLICKITM (1885) that it is greatly diminished by pre- sence of sodium sulphate. But these facts seem to afford no aid in separation of clean sulphate of quinine for weight, unless by a resort to washing with saturated solution of quinine sulphate and a correction proportional to the drying-loss. Experimental results are given by PAUL, 1877. KOPPESCHAAR (1885: Ztits. anal. Chem., 24, 362) infers that quinine and cinchonidine unite in a compound which is readily soluble in ether. 2 A note on the history of the test is given under Quinine, g. '1882: Ephemeris, i, 111. SEPARATION OF QUININE. 117 seems only practicable, in a general way, to reach near approxi- mations by some method which is simple and easy of applica- tion " : " Into the Hask containing the total alkaloids [from 5 grams bark, or 10 grams if poor in alkaloids], after these have been weighed, put first 5 grams of glass which has been ground up in a mortar to a mixture of coarse and fine powder, and then 5 c.c. of stronger ether (sp. gr. not above 0.725 at 15 C.) Cork the flask and shake it vigorously until by means of the glass all the alkaloids have been detached from the flask and ground up in the presence of the ether into fine particles. In this way the definite quantity of ether, which is large enough to dissolve all the quinine that could possibly be present, becomes entirely saturated with alkaloids in the proportion of their solubility, and the solution will necessarily embrace all the soluble ones as the quinine. Next mark two test-tubes at the capacity of 10 c.c. each, and place a funnel and a filter of 7 centimeters (2.8 inches) diameter over one of them. Wet the filter well with ether, and then pour on to it the mixture of alkaloids, ether, and glass from the flask. Rinse the flask out two or three times on to the filter with fresli ether, and then wash the filter, and percolate the glass, with fresh ether, applied drop by drop from a pipette, until the liquid in the test-tube reaches the 10 c.c. mark. Then change the funnel to the other test-tube, and continue the washing and percolation with ether until the mark on the second test-tube is reached by the filtrate. Pour the contents of the two test-tubes into two small tared capsules, evaporate to a constant weight, and weigh them. The first capsule will contain what may be called the ether-soluble alkaloids. Subtract from the weight of these the weight of the residue in the second capsule, and the re- mainder will be the approximate weight of the quinine extracted from the 5 grams of bark." These weights multiplied by 20 will give the percentage of ether-soluble alkaloids and of quinine." It is here understood that the terms " ether-soluble alkaloids " and " quinine " have a conventional meaning. And the conclusion is adopted that the quinine is all or nearly all obtained in the first 10 c.c. of filtrate, while of the less soluble alkaloids nearly equal quantities are obtained in the first and the second 10 c.c. of filtrates. Therefore the subtraction of the weight of the second residue from the weight of the first will give an approxima- tion to the weight of the quinine. Separation of Quinine (I. A, 4) from other cinchona alka- loids by solution in excess of ammonia, after crystallization of the sulphate. An adaptation of Kernels volumetric method. ii8 CINCHONA ALKALOIDS. More fully studied for Cinclionidine than for quinidine or cin- chonine. 1 Application to a Precipitate or Residue of Qui- nine with S7nall proportions of Cinchonidine (I. B, 2). The precipitate or residue is dried finally on the water-bath to a constant weight, and a weighed quantity, from 3 to 5 grams, of the dried alkaloids is taken. The alkaloids are treated with warm dilute sulphuric acid added with a little hot water to make the reaction just distinctly acidulous to litmus-paper, and retain this reaction after the alkaloids have been thoroughly saturated, when the mixture is exactly neutralized by adding dilute am- monia-water, and made up at temperature near 100 C. to a number of c.c. equal to 14.5 times the number of grams of dried alkaloid taken. The container is now placed in a bucket of water at about 15 C., along with a bottle of Standard Quinine Sulphate Solution (see Index) and a bottle of ammonia- water of sp. gr. 0.920, and the same temperature maintained for an hour or more, and adjusted at 15 C. near the close of this time. The two alkaloidal solutions are now filtered through dry filters, and the filtrates received in portions of 10 c.c. each, in test-tubes the standard quinine filtrates on one side, and the filtrates from the alkaloids to be estimated on the other side. The filtrates are titrated, in repeated trials, by adding ammonia from a burette (registering -fa c.c.), until, on gently inclining or rotating the test- tube while it is closed by the finger, the precipitate at first formed is just redissolved. Should the first 10 c.c. of filtrate under estimation require more than about 4.8 c.c. of the ammonia (0.920), after deducting the c.c. taken for 10 c.c. of the standard quinine, then a 10 c.c. filtrate under estimation should be diluted, by addition of the standard quinine filtrate, to 2, 3, or 4 times the 10 c.c. volume (20, 30, or 40 c.c.), and portions of 10 c.c. of this di- luted filtrate tested. The results of these tests, after deducting the average c.c. of ammonia for 10 c.c. of standard quinine iiltrate, are multiplied by 2, 3, or 4, to give the proper quantity of am- monia for 10 c.c, of the filtrate under estimation. Taking now the mean of the several titrations for 10 c.c. filtrate under esti- mation, after deducting the mean of titrations of standard qui- nine filtrate, each 0.32 c.c. = 0.1 per cent, cinchonidlne in the mixed alkaloids estimated. SEPARATION OF CINCHONIDINE (II. A, 1) from other cinchona alkaloids in general, after removal of quinine, ~by precipitation 1 KEENER, 1862. Improved in 1880: Zeitsch. anal. Chem., 20 ,/ifl; Archiv d. Phar., [3], 16, 186-285; 17, 488-454; Jour. Chem. Soc., 40, 63 in this work, under Quinine, g, "Kerner's Test." SEPARATION OF CINCHONIDINE. 119 with normal tartrate. The quinine may be removed (1) by crystallization as a sulphate (p. 114), or (2) by solution in ether (p. 116). For the purpose of an estimation, a deduction of the quantity of quinine from the quantity of both quinine and cin- chonidine is quite sufficient. To this end the following direc- tions of MUTER, 1 here slightly varied, serve well : The quinine is separated and estimated as crystalline sulphate (p. 114), A weighed portion of the mixed cinchona alkaloids is dissolved with hydrochloric acid enough to make the solution only slightly acid 2 to test-paper, and as concentrated as compa- tible with solution at 38 C. (or 100 F.) 3 The solution is made exactly neutral by adding sodium hydrate dilute solution, an ex- cess of the precipitant, a saturated solution of tartrate of potas- sium and sodium (Kochelle salt) is added, and the mixture kept at 15 C. (59 F.) for an hour, stirring frequently with a glass rod. The precipitate is collected on a pair of filters as small as practicable and previously (dried and) counterbalanced with each other, and is washed with, say, 100 c.c. of water at 15 C., the filtrate and wash- ings being received in a graduated measure. The precipitate is dried at 104 C. (or at 220 F.) and weighed, using the outer filter as a tare. For each c.c. of the total filtrate 0.00083 is added (MUTEK) to the weight of the precipitate. The weight of anhydrous quinine sulphate is multiplied by 0.9151, or the weight of anhydrous quinine is multiplied by 1.231, to obtain the weight of anhydrous quinine tartrate, which is deducted from the weight of the precipitate. The remainder is the weight of anhydrous cinchonidine tartrate (Ci9H 22 N 2 O) 2 C 4 H 6 p 6 , which, multiplied by 0.796T, gives the weight of cinchonidine. (For following separation of remaining alkaloids see p. 120). /Separation of Cinchonidine (II. A, 2) ~by precipitation as tartrate, followed by removal from Quinine. This plan differs from the preceding only in the order of the successive steps. In precipitating first as tartrate, in case of Commercial Quinine Sulphate, DE YKIJ (1884) directs to take 5 grams of the salt, in 200 c.c. boiling water, and add 5 grams of Kochelle salt previously dissolved in very little boiling water. After 24 hours collect on a filter, wash with the smallest quantity of water, and dry in the 'ISSO: Analyst, 5, 224. 2 Muter dissolves the mixed alkaloids in absolute alcohol, divides in two equal portions, taking one portion for quinine as a sulphate. The portion for cinchonidine is made just acid with hydrochloric acid, the alcohol evaporated off, and t he residue dissolved in least quantity of water at 100 F. 3 If the total alkaloids contain resins, kinic acid, etc , filter through a small filter, wash with as little dilution as possible, and if necessary concen- trate. 120 CINCHONA ALKALOIDS. air. KOPP states that a double normal tartrate of quinine and cinclionidine crystallizes with 1 molecule of water. HEILBIG (1880), following De Vrij, separates cinchona alkaloids in gene- ral, by initial precipitation of tart rates, as follows : 2 grams of the mixed alkaloids are dissolved as acetates in 30 c.c. of water, and the solution mixed with 1 gram Rochelle salt and well stirred. The precipitate is washed with care to avoid its solution, and dissolved in 90 per cent, alcohol acidulated with 1.6 per cent, of sulphuric acid, and herapathite is formed (as directed under Quinine, /'). The filtrate is treated with potassium iodide for precipitation of quinidine. The filtrate from the latter is treated with soda, and the resulting precipitate, dried, is exhausted by absolute ether for removal of amorphous alkaloids, the remainder being cinchonine. .for separation of Cinchonidine, Quinidine, Cinchonine, and Amorphous Alkaloids from each other, after the estima- tion of Quinine, the directions of DE VRIJ are as follows : " Two grams of the pulverized mixed alkaloids are dissolved in weak Hydrochloric acid to obtain a slightly alkaline solution measur- ing 70 c.c. By adding 1 gram of Rochelle salt to this solution," heating, cooling, stirring, and setting aside, as above indicated, " the tartrates of quinine and cinclionidine are separated ; these are collected on a filter, washed with a little water, and dried on a water-bath. One part of these tartrates represents 0.80844 of quinine and cinchonidine : from the amount of these alkaloids thus found the amount of quinine already ascertained is sub- tracted, the remainder representing the cinchonidine present." " In the filtrates from the tartrates, quinidine, if present, is pre- cipitated by a concentrated solution of potassium iodide [compare under Quinidine, d and f~\ ; one part of the dried hydriodide re- presents 0.86504 part of crystallized quinidine [0.7175 part of anhydrous quinidine]." " The remaining solution is treated with caustic soda, and the precipitate (if any) washed with ether. The residue represents the amount of cinchonine (compare under Cin- chonine, /*)." " Finally, by distilling the ether from the wash- ings can be ascertained the amount of amorphous alkaloid, which often, in the case of analysis of Indian barks, contains traces of quinamine" The directions of J. MUTER,' for separation of Quinidine, Cinchonine, and Amorphous Alkaloid, taking the filtrate from Cinchonidine and Quinine tartrates (see p. 119), are as follows : "The filtrate from the tartrate is concentrated to its original 1 1880 : Analyst, 5, 224. ROTATORY POWER. 121 volume [that before the washing of the precipitate is probably intended], cooled, rendered just faintly acid by a drop of dilute acetic acid, and excess of saturated solution of potassium iodide is added with constant stirring. After an hour or so at 15 C. [compare under Quinidine, f\ it is collected like the cinchoni- dine, and treated in every respect the same, and weighed, and the weight, having had 0.00077 added for each c c. of filtrate and washings, is multiplied by [0.7175], and result is quinidine" " The filtrate from the quinidine is made distinctly alkaline by sodium hydrate, and the precipitated cinchonine and amorphous alkaloid are filtered out in a similar manner, washed, dried, and weighed. The precipitate is then treated with alcohol of 40 per cent, to dissolve out the amorphous alkaloid, and again dried and weighed, and the difference is amorphous alkaloid, while the last weighing is cinchonine." But " the weight of the cincho- nine and amorphous alkaloid together must have deducted from it 0.00052 for each c.c. of the filtrate from the quinidine hydrio- dide, and 0.00066 for each c.c. of the filtrate from the cinchoni- dine tartrate, and the balance is then the true weight, which, minus the amorphous alkaloid, gives the cinchonine" KOTATOKY POWEK, OF CINCHONA ALKALOIDS. The plane of polarized light is deviated to the left by quinine and cinchonidine, to the right by quinidine and cinchonine. Further, the dextrorotatory alkaloids include diquinicine, quiiii- cine, cinchonicine, concusconine, conchairamine, chairamidine y and cinchotine ; the levorotatory alkaloids include hydroquinine, hydroquinidine, hydrocinchonidine, homoquinine, cusconine, con- chairamidine, paytine, aricine, and cinchamidine. The degree of deviation, or specific rotatory power, varies between the free alkaloid and its salts, 1 and varies with different solvents, concen- trations, and temperatures. Quinine hydrate, in alcohol 97$ vol., at 15 C., [a] D=-(145.2-0.657c 2 ) HESSE, 1875. Quinine hydrate in ether (0.7296), at 15 C., [a] D (158.7 1.911c) " " Quinine, anhyd., 5$ sol in chloroform, 15 C., [a] D=-106.6 " " Between 15 C. and 25 C., when c=3, ab- solute rotatory power falls 1.56 " " * Further as to the influence of the acids, OUDEMANS, 1883; as to influence of solvents, the same author, 1873. 2 c = concentration, or grams in 100 c.c. of solution. 122 CINCHONA ALKALOIDS. Quinine sulphate, cryst., in alcohol 80$ vol., (c=2), 15 C., [a] D= -162.95 HESSE, 1875. Quinine sulphate, cryst., in alcohol 60$ vol., (c=2), 15 C., [a] D= -166.36 .' Quinine bisulphate, cryst., in water, (c=l to 6), 15 C., \a\ D=-(164.85-0.31c) " " Quinine sulphate, anhyd., in water, (c 4), 15 C., [a] D=:-229.03 " 1880. Quinine sulphate, anhyd., in water, (c=l), 15 C., [a] T>= 232.7 DAVIES, 1885. Quinine sulphate, anhyd., in water, (c=4). 15 C., [a] D= 233.75 '. HESSE, 1886. Quinine sulphate, anhyd., in water, 17 C., [a\ D =242 17 OUDEMANS. Quinine hydrochloride, in water, (c=l to 3), 15 C.\ [a] D= (165.5 2 425c) HESSE. Cinchonidine, in alcohol of 97$ vol., (c=l to 5), 15 C., [a] D=-(107.48-0.297c), " Cinchonidine sulphate, 6 aq., in water, (c=1.06), 15 0., [a] D=-106.77 " Cinchonidine sulphate, anhyd., in 2.156$ sol. in alcohol, [a] D==:-153.95 " "With 0.40 gram of the salt, with 3 c.c. normal solution hy- drochloric acid, and water to make a volume of 20 c.c (" Con- centration A " of Oudemans) : Quinine tartrate, cryst., [a] D= 215.8 OUDEMANS. Anhyd.= 220. 07 KOPPESCHAAK. Cinchonidine tart., cryst., [a] D= 131.3. . . . OUDEMANS. Anhyd. =137.67 KOPPESCHAAK. Take 0.40 of mixed tartrates of quinine and cinchonidine (see under Separation of Cinchona Alkaloids by Tartrate, p. 119), dry at 125 to 130 C., dissolve as stated above for " Concentra- tion A," observe rotatory power (a), then, to find x = per cent, of quinine tartrate in the mixed tartrates : 220. 07 a? + 137.67 (100-) = 100a. _ 100(0-137.67) t ~220.07-137.67* For the estimation of cinchonidine in commercial quinine 1 KOPPESCHAAR, 1885: Zeitsr.h. anal. Chem., 24, 362; Jour. Chem. Soc., 49, 182. OUDEMANS, 1875: Arch, neerland. des Sci., 10, 193; Jahr. Chem., 1875, 140. Further, 1877 and 1884. RO TA TOR Y PO WER. 123 sulphate HESSE J directs as follows : 2 grams of anhydrous com- mercial quinine sulphate, or an equivalent quantity of crystallized salts, are weighed in a flask of 25 c.c. capacity, mixed with 10 c.c. of normal solution of hydrochloric acid, the flask fllled up to the graduation-mark with water, and, after the contents are thoroughly mixed by shaking, the solution is poured through a filter into the observation tube, which is 220 millimeters long and is provided with a water-jacket for maintaining a constant temperature. From 12 to 20 observations are made with this solution, at 15 C., and the mean of the different readings is taken. Let c = the observed deviation at the D line, and y = the cinchonidine sulphate. 8 Then, if the observation- tube be 220 m.m., y= (40.309 c) X 8.25. For other lengths of the observation-tube let C = the observed rotatory power, when y = (229.03 C) X 1.452. Quinidine, deviation diminishes with elevation of temperature. Quinidine hydrate, in alcohol of 97$ vol., at 15 C., [a] D=r+(236.T7 3.01c) HESSE. Quinidine anhyd., in alcohol of 97$ vol., at 15 C., [] D=+(269.57-3.428c) " Quinidine hydrochloride, in alcohol of 97$, at 15 C., [a] D=H-(212-2.562c) " Cinchonine, in alcohol, c=0.455, [a] D=-f-214.8 c=0.535, = 213.3 ci=0.560, = 209.6 OUDEMANS. Cinchonine sulphate, in water, c = 0.855, [a] D= +170 HESSE. Cinchonine sulphate, in 97$ alcohol, c = 0.374, [a] D =+193.29 " Cinchonine hydrochloride, [a] D=+(165 2.425c).. Quinicine, in 97$ alcohol with chloroform, [a] D:= -f(10.68-1.14c). Cinchonicine, in chloroform, at 15 C., [a] o=-[-46.5 . 1 1880: Liebig's Annalen, 205, 217; Jour. Chem. Soc., 40, 315. Also, 1885: Phar. .four. Trans., [3], 15, 869. 2 If a be the angle of rotation of dry quinine sulphate, b th* angle of anhy- drous cinchonidine sulphate, and c the "angle of the mixture, then if x be the quantity of quinine sulphate, and y the quantity of cinchonidine sulphate, the relative percentage of the last-named salt is expressed by the formula y = ^. For a and b Hesse has found the numbers 40.309 and 26.598; there- fore y = 40 ;ff~ C , or, taking y as percentage, y = (40. 309 -c) 7.293. On ac- lo.711 count of the common efflorescence of cinchonidine sulphate, Hesse modifies the formula to y = (40. 309 -c) 8.25. 124 CINCHONA ALKALOIDS. A single determination in a given solvent obviously cannot be used for estimation when more than two alkaloids of cinchona are present. But by use of different solvents, or different tem- peratures and concentrations, it has been proposed to undertake estimation in mixtures of three alkaloids. OUDEMANS has stated that optical estimation is practicable in the following-named mix- tures : quinine and cinchonidine ; quinine and quinidine ; quini- dine and cinchonidine ; quinine and cinchonine ; cinchonidine arid cinchonine ; quinine, quinidine, and cinchonidine ; quinine, quinidine, and cinchonine ; 'quinidine, cinchonidine, and cincho- nine ; quinine, cinchonidine, and cinchonine ; and tartrate of quinine, and cinchonidine. KOPPESCHAAR ' has advocated the su- perior efficiency of the optical way of estimating cinchona alka- loids, and DAviEs, 2 in report of the extended research already cited, expresses confidence in the optical estimation of cinchoni- dine in commercial quinine sulphate. HESSE, who engaged ex- tensively in optical researches upon the cinchona alkaloids in 1875, 3 and published an optical method of valuation of quinine sulphate in 1880, 4 in 1886 5 admits a diminished confidence in the optical method for exact estimations, and says that " up to the present moment we are not in possession of any optical test by which we would be able to determine the amount of cinchoni- dine in commercial quinine sulphate and other quinine salts with any satisfactory degree of accuracy." And u while constant rota- tory power in two successive recrystallizations of the same mate- rial [quinine sulphate] is satisfactory evidence of absence of cin- chonidine in that particular material, it is not by any means the case that the rotatory power of similar materials of different ori- gin is always the same." PAUL" has stated "that the results by the polariscope are much less trustworthy than those by other methods." KEENER 7 holds it to be manifestly impracticable to determine proportions of 1 and \\ per cent, of cinchonidine sul- phate, in mixtures of quinine sulphate, with even minute propor- tions of cinchonine and quinidine sulphates. The influence of hydroquinine salt, in the optical valuation of quinine sulphate, is emphasized by HESSE in the communication '1885: Phar. Jour. Trans., [3], 15, 809. *1885: Phar. Jour. Trans., [3], 16, 358. 'Liebig'8 Annalen, 176, 203-283. *Liebig's Annalen, 205, 217-222 6 Phar. Jour. Trans., [3], 16, 818, March 27, 1886. Further, same journal, June 5, 1886. 1885: Phar. Jour. Trans., [3], 16, 361. '1880: Archiv d. Phar., [3], 16, 449. QUININE. 12$ last above cited from this author. 1 He places the rotatory power of the three alkaloids chiefly concerned in the estimation of com- mercial quinine sulphate as follows. The conditions of OUDE- MANS (p. 122) are adopted : For concentration A : Quinine tartrate (a) D = 216.6 Cinchonidine tartrate 134.6 For concentration B : Quinine tartrate 212.5 Hydroquinine tartrate 176.9 Cinchonidine tartrate 132.0 Oudemans's own results were (see p. 122) : For concentration A : Quinine tartrate (a) D = 215.8 Cinchonidine tartrate 131.3 For concentration B : Quinine tartrate 211.5 Cinchonidine tartrate 129.6 QUININE. Chinin. C 20 H 24 1^ 2 O 2 324. Crystals of full hy- dration, C 20 H 24 N 2 O 2 .3H 2 O==378. Eational Formula, p. 98. Proportion* in Cinchona Barks, p. 96. Accompanying Natural Alkaloids, p. 90. Methods of quantitative separation from Cin- chona Bark, p. 102 ; from other Cinchona Alkaloids, p. 113. Means of Distinction from other Cinchona Alkaloids, schedule, p. 100. Microscopic identilication, p. 101. Optical Rotation, p. 121. Crystallization and Heat-Reactions of the free alkaloid and its salts, p. 126. Solubilities of the alkaloid and of its salts, p. 128. Physiological effects, p. 127. Quinine is recognized by the fluorescence of its sulphate solu- tion (d\ its bitterness (5), and the sparing solubility of its sul- phate in water (c). It is identified, further, by the thalleioquin test, the agreement of various reactions, and the formation of herapathite (d). The separation of quinine from other cinchona alkaloids is indexed at p. Ill ; from the bark, given on pp. 102 to 111 ; from impurities of its commercial salts, and from various common alkaloids, also from Citrate of Iron, and from Coated Pills, page 134. Means of separation are noted under e. Quinine is estimated, as stated under g, by weight of the free alkaloid, by weight of the sulphate, by weight of the iodomer- curate, by titration with Mayer's solution, and by weight of crys- tallized herapathite. The impurities and deficiencies of quinine 1 For a brief summary of the claims of De Vrij and Hesse see Am. Jour. Phar., 1886, Aug., 58, 389, editorial. Respecting optical estimations of qui- nine, treating the tartrates of the alkaloids, a paper is presented by D. HOOPER, Ootacamund, India, 1886: Phar. Jour. Trans., [3], 17, 61. 126 CINCHONA ALKALOIDS. salts (under g) are chiefly the other cinchona alkaloids, and varied quantities of water. The other cinchona alkaloids are subject to test by Kerner's method, qualitative or quantitative, and given for salts other than sulphate, the free alkaloid, the bisulphate, and for effloresced salts. Tests are given by Hesse's method, and by the directions of the pharmacopoeias of the dif- ferent nations. Concerning Liebig's test, and standards of water of crystallization, a full discussion is included under g. a. Free quinine usually appears in an amorphous or curdy or minutely crystalline white powder, or in crystals slightly efflo- resced. The trihydrate (3H 2 O) forms needles, sometimes long and silky. Crystallizing under the microscope, four- sided prisms are obtained. The precipitate by alkalies from aqueous solution of quinine salts is at first amorphous and anhydrous, but gradu- ally assumes crystallization as the trihydrate. From warm, dilute alcoholic solution anhydrous crystals have been obtained (HESSE, 1877). From ether, and most solvents other than water and alco- hol, crystals are never obtained. A dihydrate (2H 2 O) and a crys- tallizable monohydrate (H 2 O) have been reported ; also an amor- phous hydrate with 9H 2 O. The precipitate by ammonia, dried in the air at ordinary temperature, and the residue from solution in ether dried in the same way or over sulphuric acid, retain one molecule of water (FLETCHER, 1886). The trihydrate, nearly permanent in the air, loses all but about one molecule of the water slowly in the desiccator, quickly on the water-bath. All hydrates lose water gradually in warm temperatures, and on the water-bath quickly lose all but four or five per cent, (about one molecule) of the water, which is very slowly expelled (A. N. PALMER, 1876). At about 120 C. (248 F.) a constant weight of anhydrous alkaloid is promptly obtained. The trihydrate melts at 57 C. (134.6 F.) ; the anhydrous alkaloid melts, without loss, at 177 C. (350.6 F.) (HESSE, 1877), cooling to an opaque, crystalline mass permanent in the air. Strongly heated above the melting point, an amorphous, not crystalline sublimate is obtained. Crystallisation and heat- reactions of salts of quinine. 1 Quinine sulphate forms fragile, filiform crystals on the mono- clinic system, with 7H 2 O (KERNER, 1880) or 8H 2 O (HESSE, 1880). (See " Water of" Crystallization," etc., under g.) The crystals are efflorescent. The hydration is reduced, slowly in ordinary air, promptly at 50 to 60 C., to 2H 2 O. The remain- ing water is expelled slowly at 100 C., or, by three hours' dry- 1 For chemical formulaB see " Solubilities," p. 129. QUININE. 127 ing in a water-oven (H. B. PARSONS, 1884), more quickly at 112 to 115 C. The anhydrous salt recovers the 2H 2 O by exposure to the air. At or above 100 C. the salt soon begins to suffer alteration ; at about 160 C. it exhibits a greenish phosphor- escence, and above this temperature it melts, with conversion into quinicine sulphate, but without loss of weight. The salt is very slowly affected by the light. On ignition it burns very slowly, leaving no residue after complete combustion. Quinine bisulphate forms orthorhombic four-sided . prisms, or needles, sometimes nodular crystals (7H 2 O), efflorescing in the air, more rapidly in warm air, to 1H 2 O, and becoming anhydrous at 100 C. It melts in a glass tube at 80 C. (Ph. Germ.) When anhydrous it melts at or below 100 C. "At 135 C. (275 F.) it is con- verted into bisulphate of quinicine " (U. S. Ph.), this conversion beginning at the melting point, also by exposure to sunlight, and being attended with a yellowish tinge. There is a doubly acid salt, crystallizable, with 7H 2 O, in prisms. Quinine hydrobro- mide crystallizes in lustrous needles (H 2 O), " permanent in the air but readily efflorescing at a gentle heat" (U. S. Ph.), and becomes anhydrous on the water-bath. Quinine hydriodide, normal, is crystallizable in light yellow needles, instable, easily altered to a soft, resinous mass. Quinine nitrate crystallizes with difficulty in very oblique prisms (H 2 O), easily melted to an oily mass, and becoming anhydrous at 100 C. Quinine val- erianate crystallizes in pearly, triclinic crystals (H 2 O), perma- nent in the air, melting at about 90 C., becoming anhydrous at 100 C., at which temperature it also begins to lose valerianic acid. Quinine normal tartrate, H 2 O, becomes anhydrous at 100 C. Quinine oxalate, normal, crystallizes with 6H 2 O, in very fine needles. J. Quinine is odorless, and has a pure bitter taste of much intensity. The persistence and intensity of the bitter taste of quinine salts is in proportion to their solubility as brought in contact with the tongue. Of ordinary forms administered the tannate is the least and the free alkaloid next least bitter, the sulphate being less bitter than the bisulphate, hydrobromide, or hydrochloride. Quinine is poisonous to the lower forms of ani- inal life, in this effect being surpassed among vegetable poisons only by such as strychnine and morphine (Bmz). For frogs the fatal dose is 0.05 to 0.1 gram (} to 1| grain) internally, or about 0.0025 (| grain) subcutaneously. For dogs about "0.12 gram per kilogram (-J grain per pound) of body-weight proves fatal (BERNATZIK, 1867). Infusoria and bacteria are destroyed with 128 CINCHONA ALKALOIDS. somewhat concentrated solutions of quinine salts, quite variable strengths being required for different infusoria. Quinine is antiseptic, hindering or stopping the alcoholic, lactous, butyrous, amygdalous, and salicylous fermentations (Bmz, " Husemann's Pflanzenstoffe," 1884), not the digestive action of pepsin. Qui- nine is excreted in the urine to the extent of 70 to 96 per cent, of the amount taken. It appears in the urine as early, fre- quently, as one hour, and usually disappears as soon as forty- eight hours, after ingestion (KEENER, JURGENSEN, and FRAU). Quinine is found in the liver. In some small part, also, it suf- fers conversion in the system into amorphous quinine [di- quinicine?], and an oxidation product, Dihydroxyl-quinine (C 20 H 24 N 2 O 4 ) (KERNER), or, according to SKRAUP, Chitenine (C 19 H 22 N 2 O 4 ). Kerner states that the physiological action of the oxidized product is much weaker than that of quinine. Chi- tenine is formed by action of permanganate on quinine, is insolu- ble in ether, fluoresces, and gives the thalleioquin reaction. G. Solubilities. Quinine is sparingly soluble in water; quite freely soluble in alcohol, ether, chloroform, amyl alcohol ; mode- rately soluble in water of ammonia, benzene, glycerine ; and sparingly soluble in petroleum benzin. The alkaloid trihydrate is soluble in 1670 parts water at 15 C. (HESSE), in 1428 parts water at 20 C. (SESTINI, 1867), in 760 parts boiling water (EEG- NAULD, 1875), in 902 parts boiling water (SESTINI), in six parts of ordinary alcohol at 15 C., in ly^ parts absolute alcohol (REG- NATJLD), in 2 parts boiling alcohol of 90$, in " about 25 parts of ether" (U. S. Ph.), in 22 parts of ether at 15 C. (REGNAULD), in u about 5 parts of chloroform " (U. S. Ph.) The anhydrous alkaloid is soluble in 1960 parts of water at 15 C. (HESSE), in about the same proportion of ether required for the hydrate (HESSE), in (near) 2 parts chloroform (PETTENKOFER, 1858), in 200 parts benzene at 15 C. or 30 parts boiling benzene (OuDE- MANS, 1874). Crystals, mostly needle-form, can be obtained from nearly all solutions. From benzene, crystals of C 20 H 24 N 2 O 2 -|-C 6 H 6 are obtained (OUDEMANS). Solubility in ether is diminished by pre- sence of other cinchona alkaloids (PAUL, 1877). Quinine has a decided alkaline reaction, promptly shown upon test-papers in the aqueous solution. The normal salts of the stronger acids are neutral to litmus, the sulphate of manufacture not -infrequently alkaline in the least perceptible degree. Quinine salts of ordinary acids are soluble or moderately soluble in water and in alcohol, except the sulphate, which is only sparingly soluble QUININE. 129 in water. The proportion of water required for the free alkaloid at 100 C. is about that required for the sulphate at 15 C. Solubilities of quinine salts. - - Quinine sulphate, (C 20 H 24 ]^ 2 p 2 ) 2 H 2 SO 4 .7H 2 O=872, is soluble " in 740 parts of water and in 65 parts of alcohol at 15 C. (59 F.) ; in about 30 parts of boiling water, in about 3 parts of boiling alcohol, in small proportions of acidulated water, in 40 parts of glycerine, in 1000 parts of chloroform, and very slightly soluble in ether " (IT. S. Ph.) Its solubility in water is decreased by presence of am- monium sulphate (CABLES) or sodium sulphate (SCHLICKUM, 1885) ; in chloroform is increased by presence of cinchonine or quinidine -sulphate. From acidulous aqueous solution it is sparingly dis- solved by amyl alcohol (BAKFOED). In alcoholic solution it is pre- cipitated by adding ether. Qjuinine Usulphate, C 20 H 24 N 2 O 2 H 2 SO 4 .7H 2 O=:548, is soluble "in about 10 parts of water (with vivid blue fluorescence) and in 32 parts of alcohol, at 15 C. (59 F.) ; very soluble in boiling water and in boiling alcohol " (U. S. Ph.) It has a strongly acid reaction. The doubly acid sulphate, C 20 H 24 N 2 O 2 (H 2 SO 4 ) 2 7H 2 O, is freely soluble in water and in alcohol. Quinine hydrobromide, C 00 H 04 N 2 O 2 HBr . H 2 O =422.8, is soluble " in about 16 parts of water and in 3 parts of alcohol, at 15 C. (59 F.) ; in 1 part of boiling water and less than 1 part of boiling alcohol ; in 6 parts of ether, in 12 parts of chloroform, and moderately soluble in glycerine" (U. S. Ph.) Quinine hydrochloride (muriate), C 22 H 24 N 2 O 2 HC1.H 2 O==378.4, is soluble "in 34 parts of water, and in 3 parts of alcohol, at 15 C. (59 F.) ; in 1 part of boiling water and very soluble in boiling alcohol ; when rendered anhydrous it is soluble in 1 part of chloroform " (U. S. Ph.) In 9 parts of chloroform (Hager's " Commentar " ). Normal quinine hydriodide, instable, is more soluble than the sulphate. Quinine valerianate, C 20 H 24 1S[ 2 O 2 C 5 H 10 O 2 . H 2 O=444, is soluble in about 100 parts of water and in 5 parts of alcohol, at 15 C. (59 F.), ... and slightly soluble in ether " (U. S, Ph.) -Quinine tannate? amorphous, is but very little soluble in cold water (nearly tasteless), but is soluble in alcohol and slightly soluble in ether, and by long digestion with water is converted into soluble quinine gallate (LINTNER). Qui- nine tartrate, normal (C 20 H 24 N 2 O 2 ) 2 C 4 H 6 O 6 . H 2 O, is soluble in 910 parts of water at 10 C., much more soluble in hot water and in alcohol (HESSE, 1865). Quinine oxalate, (C 20 H 22 N 2 O 2 ) 2 1 JOBST, 1878. Fluckiger's "Phar. Chemie,"425. Hager's " Phar. Praxis," iii. 291. Produced of very variable composition and properties. Ascribed formula, C2oH 2 4N 2 02(Ci4Hio09)3=25 per cent, quinine. Jobst prepared it, 31 percent, quinine; and found it in commerce from 7 per cent, to 22 per cent, quinine. 130 CINCHONA ALKALOIDS. H 2 C 2 O 4 .6H 2 O, requires 898 parts of water at 10 C. for solu- tion ; 1446 parts at 18 C. (SHIMOYAMA, 1885). d. Fluorescence. In general, quinine salts with inorganic acids containing oxygen exhibit blue fluorescence in their aqueous solutions. The hydracids of chlorine, bromine, etc. , the cyanogen hydracids, and thiosulphuric acid, in union with quinine, do not give fluorescence. By adding sulphuric acid the fluorescence is obtained with all the salts in aqueous solution. But the hydra- cids, if present, in proportion to their quantity diminish the reac- tion. Alcoholic solutions show little fluorescence ; solutions in ether, chloroform, etc., none at all. The bisulphate fluoresces much more strongly than the normal sulphate, in solutions of equal strength, and the fluorescence of a neutral solution of the sulphate is much intensified by acidulating with sulphuric acid. To obtain the full delicacy of the reaction, put the solution in a test-tube or beaker of such width that a depth of at least two inches is obtained. Place over a black ground, in a strong light falling horizontally from one direction, observing from above, comparing with a like column of distilled water, and, if neces- sary, shading the eye from the direct light and shading the liquid from the lateral light. Greater intensity is attained by throwing the light from a lens in a pencil upon the liquid. 1 So observed, 0.00005 gram quinine, in 5 c.c. acidulous solution, gives distinct fluorescence, and this (1 in 100000) is not the limit of dilution (BARFOED, 1881). The fluorescence of quinine is shared by qui- nidine, and by diquiniciiie, hydroquinine, hydroquinidine ; not by cinchonidine nor by quinicine. Thatteioquin test. Treated in a white porcelain dish with fresh chlorine-water or bromine-water, not in too great excess, or well diluted, and then with ammonia to just effect an alka- line reaction, a solution of quinine gives a green precipitate, thalleioquin, dissolving to a green solution by adding a further excess of ammonia. In more dilute solutions a precipitate is not obtained at all, but a green liquid. Bromine gives with dilute solutions a better result than chlorine (ZELLEK, 1880); an exces- sive action of either is to be avoided. According to BAKFOED a fine reaction is given by 0.001 gram of quinine, in 5 c.c. of water acidulated with sulphuric acid, treated with 10 drops of very weak bromine- water or of fresh chlorine- water, and then with 2 drops of ammonia-water: but with 0.0005 gram, in 5 c.c., 2 drops J For more minute examination see STOKES, 1853; H. Morton, 1871; " Watts's Diet.," 3, 634; 8, 1193. QUININE. 131 weak bromine -water and 1 drop ammonia-water, the limit is readied. TRIMBLE (1877) lias used the reaction for a colorome- tric method, and prepared a standard green solution by propor- tions of 0.01 quinine or quinine salt in 5 c.c. of fresh chlorine - water, adding 10 c.c. of ammonia-water and diluting to 100 c.c. If the green ammoniacal solution be just neutralized with acid a blue tint is obtained, and, by acidulating, a violet to red color, returning to green again when ammonia is added in excess. If ferricyanide of potassium be added after the chlorine or bro- mine addition as above, and then ammonia barely enough for an alkaline reaction, a red color is obtained. Froehde's reagent, with dry quinine, gives a slight green color (DRAGENBORFF). The thalleioquin test of quinine is shared by quinidine, diquini- cine, and quinicine, also by hydroquinine and hydroquinidine, but not by cinchonidine nor cinchonine. The alkali hydrates precipitate quinine from solutions of its salts, the precipitate becoming slowly crystalline (see $), and being quite readily soluble in excess of ammonia, and somewhat soluble in excess of ammonium carbonate, not of the fixed alkali hydrates, or only very slightly by potassa. Tartaric acid pre- vents the precipitation in solutions more dilute than about 1 to 300 ; and ammonium chloride increases the solubility of the pre- cipitate. In free ammonia the quinidine and cinchonidine pre- cipitates are less soluble than that of quinine, and the cinchonine precipitate is but very slightly soluble. The alkali carbonates, and, more slowly, the bicarbonates, precipitate quinine, insoluble or, with bicarbonates, but slightly soluble in excess. Herapathite test. Herapathite (HERAPATH, 1852) is one of the iodosulphates of quinine. Its formula (JORGENSEN, 1876) is (C ?0 H2 4 N" 2 p a ) 4 (H 2 SO 4 ) 3 (HI) 2 I 4 .(H 2 O)n. Dried at 100 C , it contains 55.055 per cent, anhydrous quinine. It crystallizes in plates, either rectangular or rhombic, of six or eight sides. By reflected light the crystals are very lustrous, or iridescent emerald- green ; by transmitted light they are dichroic, in the direction of one axis nearly transparent, but when certain axes are super- imposed they are nearly opaque. A play of dark and light shades is obtained with crystals of microscopic size floating in a drop of liquid undei? the cover-glass. The large crystals have the optical powers of tourmalines, but in greater intensity. Herapathite is at first nearly insoluble in cold water and soluble in 1000 parts hot water, but is decomposed by water with formation of quinine bisulphate and hydriodide. It dissolves in 50 parts boiling alco- hol of 85$ by weight; in 650 parts of cold alcohol of same 1 32 CINCHONA ALKALOIDS. strength. In 800 parts of 90$ alcohol at 10 C. In 751 parts of 92$ alcohol at 24.5 C. (76.1 F.) (L>E YKIJ, 1875). It is always crystallized from alcohol, usually acidulated. The large crystals of herapathite can be mechanically separated from amorphous precipitate of other cinchona alkaloids. DE YKIJ states (1882) that the best reagent for the quali- tative recognition of crystallizable quinine, when in a mixture of cinchona alkaloids, is the iodosulphate of chinoidine, pre- pared as directed (under f) for quantitative uses. This is added to a solution of 1 part of cinchona alkaloids dissolved in 20 parts of 92-95$ alcohol acidulated with 1.5$ of sulphuric acid, this solution being then diluted with 50 parts of alcohol. The iodosulphate reagent is added (before heating) so long as a dark brown-red precipitate is formed, when, with slight excess of reagent, the liquid acquires a yellow color. The mixture is now heated to boiling, to dissolve the precipitate^ then set aside for crystallization of the herapathite. BAKFOED (1881) dissolves alka- loid supposed to contain 0.01 gram quinine in 20 drops, or 0.01 gram quinine sulphate in 10 drops, of a mixture of 25 drops of alcohol, 30 drops of acetic acid, and 1 drop of diluted sulphuric acid, heating to boiling, and then adding 2 drops of alcoholic solution of iodine (1 to 200) and setting aside. Crystallization may begin in 15 to 30 minutes. Excess of iodine tends to produce other iodosulphates of quinine. JORGENSEN (1876) describes three classes of these : C 20 H 24 N 2 2 ) 2 (H 2 S0 4 ) (HIU C 20 H 24 N 2 3 ) 3 (H 2 S0 4 ) Olive-gray, bronze, brown, blue, and black colors are found, as well as green shades; and needles, as well as plates. The results are governed mainly by the proportions of quinine, io- dine, and sulphuric acid taken. The other cinchona alkaloids form iodosulphate precipitates, somewhat more soluble in alco- hol, and less crystallizable, than quinine iodosulphate. CHKISTEN- SEN (1881) states that cinchonidine, if present in at all large quantity, may be precipitated even by De Yrij's method with chinoidine. See also Cinchonine, d. Further citations from DE YKIJ are given under /, " Quantitative" p. 136. General reagents for alkaloids. Potassium mercuric iodide, or Mayer's solution, precipitates quinine in white flakes, appear- ing in acidulous solutions containing loss than I part of the alka- loid in 100000 (/, p. 136). Phosphomolybdate throws down quinine from acidulous solutions, the yellow- white, curdy preci- QUININE. 133 pitate being almost absolutely insoluble. 1 Iodine in potassium iodide solution causes a reddish- brown precipitate. In solutions other than that of the sulphate the precipitate is at first soft or amorphous ; in presence of sulphuric acid the precipitate ap- proaches to the composition and appearance of herapathite. See Cinchonine, d. Platinic chloride, in solutions not very dilute, a bright yellow precipitate, C 20 H 24 N 2 O 2 (HCl) 2 PtCl 4 , soluble in 1500 parts of cold water or in 2000 parts of boiling alcohol. Tannic acid, a yellow- white amorphous precipitate (see p. 48), easily soluble in warm hydrochloric acid. Picric acid, in satu- rated aqueous solution, a yellow amorphous precipitate, soluble in alcohol, from which it crystallizes. Potassium sulphocyanate, in concentrated solutions, a white precipitate, more soluble than the sulphate (HESSE), used in microchemical examination, p. 101. Sulphates give a precipitate in neutral solutions of hydro- chloride and hydrobrornide of quinine, if not diluted to the ex- tent of the solubility of quinine sulphate. Concentrated sul- phuric acid causes no color; Froehde's reagent a greenish color. Potassium iodide, in neutral solutions moderately dilute, does not precipitate quinine salts (separation from quinidine). A saturated solution of quinine sulphate is not affected. The slightest acidulation, such as may take place in the stomach, may result in the liberation of iodine and the formation of insoluble quinine iodides resembling herapathite. Normal tartrates, as potassium sodium tartrate, precipitate moderately concentrated solutions of quinine salts, the normal tartrate of quinine being a little more soluble (c) than that of cinchonidine, and much less soluble than those of quinidine and cinchonine (to be observed in cinchonidine separation by tartrate). e. Separations. All the cinchona alkaloids, in aqueous solu- tions of their salts, or other solutions of free alkaloid, are evapo- rated to dryness at 100-125 C. without loss. From substances insoluble in ether, chloroform, or amyl alcohol, quinine is sepa- rated by action of these solvents, none of which dissolves salts of quinine, except chloroform very slightly. Benzene in sufficient quantity dissolves quinine, as does aqueous ammonia. Methods of separation of quinine from Cinchona Bark are given, pp. 102 to 111; from other Cinchona Alkaloids, index at p. 111. From Morphine and from Strychnine quinine is pretty nearly sepa- rated by its solubility in ether, less fully separated by its solu- bility in ammonia. In the sulphates quinine is approximately 'The author, 1877: Am,. Jour. Phar., 49, 483. 134 CINCHONA ALKALOIDS. separated from Morphine and from Atropine by the differences of solubility in water. From Salicin it is well separated, as free alkaloid, by its insolubility in water. From Citrate of Iron and Quinine. The assay method of the U. S. Ph. is as follows : " The salt contains 12 per cent, of dry quinine. It may be assayed as follows : Dissolve 4 grams of the scales in 30 c.c. of water, in a capsule, with the aid of heat. Cool, and transfer the solution to a glass separator, rinsing the capsule ; add an aqueous solution of 0. 5 gram of tartaric acid, and then solution of soda in decided excess. Extract the alka- loid by agitating the mixture with four successive portions of chloroform, each of 15 c.c. Separate the chloroformic layers, mix them, evaporate them in a weighed capsule, on the water- bath, and dry the residue at a temperature of 100 C. (212 F.) It should weigh 0.48 gram." The Br. Ph. process is as follows: " Fifty grains [or 4 grams] dissolved in a fluid-ounce [or 35 c.c.] of water and treated with a slight excess of ammonia gives a white precipitate, which, when dissolved out by successive treat- ments of the fluid with ether or chloroform, and the latter evapo- rated, and the residue dried until it ceases to lose weight, weighs eight grains [or 0.639 gram]." Mr. J. C. FALK' advises to add 1 gram of tartaric acid in the U. S. Ph. process, as he found the 0.5 gram insufficient to keep the iron in perfect solution. The four portions of chloroform are often insufficient. The solvent should be applied till a portion ceases to give test for quinine. Analysts often find it difficult to recover the entire quantity added. The use of a continuous extraction-apparatus for liquids is desirable. Mr. Falk recovered 11.925 from the addition of 12. The recovered alkaloid is tested most readily by solubility in ether, more certainly by the application of the Ammonia Test to free alkaloid (p. 139). Where the ammonia test is the official standard for quinine hydrate and its several salts, it is the just and indisputable stan- dard for the alkaloid obtained from all preparations of quinine, such as pills, scales, elixirs, etc. Of 34 samples of citrate of iron and quinine assayed by Dr. DAVENPORT, State Analyst of Drugs in Massachusetts, 2 by the U. S. Ph. method, 85 per cent, fell below the pharmacopoeial requirement, though the greater proportion were not in the phar- macopoeial form of the preparation. From Coated Pills of Quinine Salts. The following method 1 1884: Am. Jour. Phar., 56, 316. 2 " Fifth Annual Report State Board of Health," etc., Mass., 1884, p. 162. QUININE. 135 contributed by HENRY B. PARSONS,' and verified by use in his constant practice, is confidently recommended : Take a sufficient number of pills to represent 20 or 40 grains of sulphate of qui- nine ; a treat, in a very small Wedgewood mortar, with 5 c.c. cold water until the coating dissolves and a smooth and uniform paste is obtained ; add 2 grams (31 grains) of freshly slaked lime in powder ; mix thoroughly and dry the mixture slowly in the mor- tar by a steam or water bath. The dry mass is to be finely pow- dered and transferred 3 to a Tollens apparatus for continuous percolation, 4 and thoroughly extracted with stronger ether. Eva- porate the ethereal solution in a weighed flask, dry for one hour at 125 C., and weigh as anhydrous quinine. Grams of anhy- drous quinine X 20.7673 = grains of quinine sulphate (7H 2 O). f. Quantitative. Gravimetric estimation as free alkaloid. Quinine is frequently estimated by weighing the residue from a solution of the separated alkaloid in ether, chloroform, or amyl alcohol a method without objection (see 0, p. 134). The residue is preferably dried first at a very moderate heat or over sulphuric acid to avoid melting (a), and finally at about 120 C., arid cooled in a desiccator. The objection to precipitation for weight is the loss by solubility in water. Sodium hydrate is without ob- jection as a precipitant. In precipitating quinine sulphate aci- dulate solution with sodium hydrate, and washing on the filter until the washings gave no cloudiness with barium chloride, a loss of 11.6 per cent, of the quinine was sustained. The solu- bility in sodium sulphate solution is about the same as that in pure water. Dry quinine, washed on the filter, ordinarily loses about 0.0002 gram per c.c. of wash- water; but a watery filtrate fully saturated with quinine will contain about 0.0006 1 1883: New Rem., 12, 67; Proc. Am. Pharm., 31, 270. 2 The smaller number is sufficient if manipulations are made with care and the balance is sensitive to tenths-milligram. 3 By use of a small steel spatula. The mortar then rinsed with a little of the ether. 4 In absence of a Tollens apparatus good results may be obtained by a very careful operation on an aliquot part of the solution as follows : Transfer the dry mass to a small, flat-bottomed flask; measure in an exactly taken volume of stronger ether, [stopper, and weigh] and agitate the stoppered vessel, occasion- ally, while it stands for 12 hours or more. [Weigh again and add ether to restore the loss if any has occurred.] By use of a pipette measuring accurately [and agreeing with the measure by which the ether was taken], take off from the clear ethereal solution an aliquot part by volume of the ether taken, and evaporate as directed for the percolate. Chloroform does not work as well as ether as a solvent. With a faithful execution of the process the loss is not over | per cent. In answer to the opi- nion of MASSE (1885) that quinine suffers loss by action of lime at 100 C., see PASSMORE (1885). 136 CINCHONA ALKALOIDS. gram per c.c. of the liquid (c). 1 The precipitate is preferably dried first at a gentle heat, and at last at about 120 C. (248 F.), as the last molecule of water is difficult to expel at 100 C. Heat to about 170 C. is borne without loss of alkaloid. The dried alkaloid must be cooled in a good desiccator, as it readily acquires water from the air. Gravimetric determination as crystallized sulphate, dried at 100 C. (or 115 C.) to anhydrous sulphate, or at 60 C. to- effloresced sulphate (2H 2 O), is directed under Separation of Cin- chona Alkaloids, p. 113. Gravimetric determination as quinine mercuric iodide by precipitation of the acidulated solution of the sulphate, with Mayer's solution, gives fair results. The precipitate is washed, and dried at 100 C., when 2.900 grams indicate 1.000 gram of quinine (as dried at 100 C.) 2 The composition of the precipitate is per- haps liable to variation by action of solvents, bat it is almost in- soluble in water. The precipitate by phosphomolybdate, in acid- ulated solution, may be washed, dried below 70 C.. and weighed,, when 1 gram of quinine is represented by about 3.665 grams of the precipitate, 3 the result being properly controlled by a par- allel operation upon a known quantity of pure quinine. Volumetric estimationby Mayer '# Solution. The precipitate,, as stated under d (p. 132), has very little solubility, but its com- position is probably varied by conditions of temperature, etc. According to Mayer, in dilution of 1 to 800, 1 c.c. of the re- agent = 0.0108 gram of anhydrous quinine. 4 It is advisable to> control the results by a parallel titration of a solution of quinine of known strength. Estimation in herapathite (DE YRIJ, 1882). Preparation of the Reagent, lodosulphate of Chinoidine. Of commercial chinoidine 1 part is heated on the water -bath with two parts of benzene, whereby the chinoidine is partly dissolved. The clear,, cold benzene solution is shaken with an excess of weak sulphuric acid, whereby a watery solution of acid sulphate of chinoidine is- obtained. After ascertaining in a small part of this solution the amount of amorphous alkaloid contained in it, so that its whole 1 " Laboratory Notes," by the author, 1877: Am. Jour. Phar., 49, 481; Jour. Chem. Soc., 32, 933; Jahr. Phar., 1877, 419. 2 "Laboratory Notes," by the author, 1877 (where last cited). 3 Last citation. 4 BLYTH, 1881: The Analyst, 6, 161; New Rem., II, 34; Proc. Am. Phar. r 30, 410; Jour. Chem. Soc., 40, 1176. The factor 0.0108=1 of ^J^r of the m - lecular weight, and indicates the formula (C 2 oH 24 N 2 O 2 ) 2 (HI) x , (HgI 2 ) 3 for the precipitate, but this is not supported by the gravimetric results (A. B. PRESCOTT,. 1880: Am. Chem. Jour., 2, 294; Chem. News, 45, 114). QUININE. 137 quantity in the solution may be known, the clear solution is poured into a large capsule. For every two parts of amorphous alkaloid contained in the solution 1 part of iodine and 2 parts of iodide of potassium are dissolved in water. This solution is slowly added with continuous stirring to the liquid in the cap- sule, so that no part of it comes in contact with an excess of iodine. By this addition there is formed an orange- colored, flocculent precipitate of iodosulphate of chinoidiiie, which, either spontaneously or by a slight elevation of temperature, collapses into a dark brown-red colored resinous substance, whilst the supernatent liquor becomes clear and slightly yellow-colored. This liquor which, if the direction is strictly followed, must still contain some amorphous alkaloid as a proof that no excess of iodine has been used is poured oft', and the resinous sub- stance is washed by heating it on a water-bath with distilled water. After washing, the resinous substance is heated on a water- bath till all water has been evaporated. It is then soft and tenacious at the temperature of the water-bath, but becomes hard and brittle after cooling. One part of this substance is now heated with 6 parts of alcohol of 92 to 95 per cent, on a water-bath, and is thus dissolved, and the solution allowed to cool. In cooling, a part of the dissolved substance is separated. The clear dark brown-red colored solution is evaporated on a water- bath, and the residue dissolved in 5 parts of cold alco- hol. This second solution leaves a small part of insoluble sub- stance. The clear dark brown-red colored solution obtained by the separation of this insoluble matter, either by decantation or filtration, constitutes the reagent which, under the name of " iodo- sulphate of chinoidine," Dr. De Vrij uses both for qualitative and quantitative determination of the crystallizable quinine in barks. The formation of herapathite, in the estimation, is directed by De Yrij as follows : Of the mixed alkaloids from a cinchona bark, 1 part (1 gram being sufficient) is dissolved in 20 parts of alcohol of 92 to 95 per cent., containing 1.5 per cent, of sulphuric acid (of which an excess above that for production of acid sul- phates is avoided). The resulting alcoholic solution of the acid sulphates of the alkaloids is then diluted with 50 parts of pure alcohol. From the dilute solution so obtained the quinine is precipitated at the ordinary temperature by adding carefully, by means of a pipette, the above-mentioned solution of iodosul- phate of chinoidine as long as a dark brown-red precipitate of iodosulphate of quinine (herapathite) is formed. As soon 'as all the quinine has been precipitated, and a slight excess of the re- 138 CINCHONA ALKALOIDS. agent has been added, the liquor acquires an intense yellow color. 1 The beaker containing the liquor with the precipitate is now covered by a watch-glass, and heated till the liquid begins to boil and all the precipitate is dissolved. The beaker is then left to itself, and in cooling the herapathite is separated in the well- known beautiful crystals. After twelve hours' rest [finally at 16 C.] the beaker is weighed to ascertain the amount of liquid which is necessary in order to be able to apply later the neces- sary correction; for although the quinine-herapathite is very slightly soluble in cold alcohol, it is not insoluble (d, p. 131). It is ascertained with a small portion of the solution that enough reagent has been added, when the clear liquid is poured off, as far as possible, on a filter, leaving the majority of the crystals in the beaker, which is now weighed again to ascertain the amount of liquid, which is noted down. The crystals are now dissolved to recrystallize, for removal of traces of adhering cinchonidine iodosulphate, as follows : The crystals on the filter are washed into the beaker with a little of the alcohol, and all the crystals dissolved in just enough alcohol at the boiling point. After perfect cooling [and standing at 16 C.] the weight of the beaker is taken, the crystals carefully collected on a small filter, and the empty beaker weighed again. The difference in weight will indicate the amount of liquor, which is added to that of the first liquor. The sum of the weights of the li- quor X 0.00125 = correction for solubility of the herapathite, provided the crystallization has been completed at 16 C. 2 The herapathite crystals on the filter are thoroughly washed with a saturated alcoholic solution of pure herapathite. 3 The washed crystals are drained, with tapping of the side of the funnel, the filter taken out and quickly pressed between blotting-papers, and as soon as air-dry the crystals are transferred from the filter to a fitted pair of watch-glasses, and dried on the water-bath (or at 100 C.) to a constant weight, avoiding the access of air. To the weight is added the correction for solubility, to obtain the total quinine iodosulphate, (C 20 II 24 N 2 O 2 ) 4 (II 2 SO 4 ) 3 (HI) 2 I 4 (d, 1 If cinchonidine be present in large quantity, the author states that the due control of this slight excess of the reagent requires a great deal of practical experience, and must be studied on a solution of cinchonidine itself, taken in the proportions above directed. 2 If another temperature has been employed, the solubility of the hera- pathite is to be determined by experiment at such temperature. In this the herapathite can be estimated by volumetric hyposulphite: 21.58 parts of iodine representing 100 parts of herapathite. 3 A washing-bottle containing an excess of pure crystallized herapathite in 95 per cent, alcohol may be kept ready for application. QUININE. 139 p. 132), of which one part contains 0.55055 part of anhydrous quinine. g. Tests for impurities and deficiencies. The impurities or deficiencies of quinine salts to be generally regarded are, in or- der of importance, (1) other cinchona alkaloids in excess of a proper limit, and (2) an excess or deficiency of moisture or water of crystallization, causing variableness of strength. Quinine manufacture is mainly conducted by a small number of houses of well-known standing, and the product is carried in well-regu- lated commercial channels, so that it is but little exposed to the introduction of falsifications. The one cinchona alkaloid, not quinine, most difficult for the manufacturer to remove and for the analyst to estimate, and actually present in largest proportion in the product from barks in general, is cinchonidine. In the pro- duct of the " cuprea " barks, however, another alkaloid is intro- duced, which is cupreine, or the conjugated compound of cupreme with quinine known as homoquinine, and it becomes necessary to give general attention to the possible presence of this alkaloid. The recognized tests for other cinchona alkaloids depend, in principal, upon (1) the removal of quinine as a crystallized sul- phate (KEENER, 1862), (2) the separation of the free alkaloids by ether (Liebig's test), or (3) by excess of ammonia (KEKNEK, 1862), which is used also in all tests to liberate the alkaloids from their salts. Hesse's test (1879) depends upon principles (1) and (2), as also does Paul's (1877), while Kerner's test depends upon (1) and (3). Kernels test of Quinine Sulphate provides a uniform arbi- trary measure, by certain fixed conditions, as follows : Quinine as a sulphate is macerated with water; the quantity of the water is 10 parts for 1 part of crystallized sulphate ; at whatever tempera- ture the maceration is commenced, it is invariably concluded at the temperature of 15 C., when the mixture is at once filtered ; and of the filtrate 5 c.c. (for some purposes 10 c.c.) are treated with an accurately measured volume of ammonia-water of exact- ly known strength (s.g. 0.920 or 0.960) until a clear liquid^is ob- tained, the ammonia- water being mixed at once with the filtrate by gently inclining or rotating the test-tube while this is closed with the finger. With a small and not bibulous filter 1 gram of crystallized sulphate with 10 c.c. of water will easily yield the 5 c.c. of filtrate for one ammonia test. Directions often specify 2 grams of the crystals with 20 c c. of water; and for quantita- tive titrations it becomes proper to take 5 grams of the crystal- lized sulphate with the tenfold number of c.c. of water, to pro- HO CINCHONA ALKALOIDS. vide for several parallel ammonia tests; 1 but it will be observed that the required quantity of ammonia, the index of the test, is only placed in ratio to the 5 or 10 c.c. of the filtrate [where it acts not wholly independent of variations of atmospheric temperature], and has no ratio to the quantity of quinine sulphate taken. It will be further observed that the fixed proportion of water taken in maceration, 10 parts to 1 of the salt, controls the quantity and concentration of the alkaloids not quinine. The 10 parts of water at 15 C. would dissolve 0.1 part of cinchonidine, 10$ of the quinine sulphate taken, and a smaller quantity of water would in most cases dissolve the entire quantity of alkaloids not quinine, but it is of importance that the solution of these alka- loids shall be made up to the same volume in every trial of the test. vThe quinine sulphate is in excess of saturation of this salt ; indeed, one-fiftieth as much quinine salt would suffice to more than saturate the 10 parts of water. In Kerner's method the Quinine Sulphate is readily recovered in purified form, almost without waste, and sometimes with gain, of value. Of the real quinine sulphate 99.86$ remains on the filter as Recrystalli zed Sulphate of Quinine. It is dried by pressing gently between blotting-papers and setting aside in dry air, avoiding efflorescence. The quinine dissolved by the ammonia crystallizes on evapora- tion of the latter, and this separation has been adopted in puri- fying small quantities of quinine. For 5 c.c. of aqueous solution saturated at 15 C. the volumes of ammonia-water of sp. gr. 0.92 (21.7$ NH 3 ), and of sp. gr. 0.96 (10$ NH 3 ), required to give a clear solution, are as follows (KERNEB, 1880 2 ) : 1 HAGER arranges for special tests, in different portions of the filtrate, for single cinchona alkaloids, but of these special tests the only trusty one is that for quinidine, rarely present a test made with 5 c.c. of the nitrate by adding 5 drops of a 1 to 20 solution of potassium iodide, stirring, and setting aside for crystallization of quinidine hydriodide. 2 Archiv d. Phar.. [3], 17, 444. In these experiments with alkaloids other than quinine they were sometimes added (in known quantity of their sulphates) to the 5 c.c. of the filtrate from pure quinine sulphate, but in case of cinchoni- dine it was mixed with the crystallized pure quinine sulphate for one series of trials, and in certain of the tests there was maceration with water at 60 and at 80 C. before the crystallization at 15 C. These varied conditions made little difference with the results. But when the quinine and cinchonidine sulphates have been crystallized together, previous hot digestion increases the efficiency of the separation. See YUNGFLEISCH, 1887: P/iar. Jour. Trans. [8] 17, 585; Jour, de Pharm. [5] 25, 5. The same is stated in general terms by PAUL 1877: Phar. Jour. Trans. [3] 7, 654. The author is indebted to E. A. RUDDIMAN for determinations of difference due to previous hot digestion, a report of which, will soon be presented for publication. QUININE. 141 With pure Quinine Sulphate Ammonia ofs.g.0. 960. 5.0 to 5.3 c.c. Ammonia of s.g. 0.920. 1 3.0 to 3.3 c.c. For 0.001 gram Cin- chonidine sul- phate added .... For each per cent, of Cinchonidine sulphate. . Additional to above. 0.40 c.c. to 0.44 c.c. 2.0 e.c. to 2. 2 c.c. Additional to above. 0.28 to 0.35 (av. 0.32) c.c. 1.4 to 1.7 (w. 1.6) c.c. For 0.001 gram Quinidine sul- phate 1.16 c.c. to 1.34 c.c. 0.56 c.c. to 0.78 c.c. For 1 per cent. Qui- nidine sulphate. . For 0.001 gram Cinchonine sul- phate 2 5.8 c.c. to 6.7 c.c. 0.62 c.c. to 0.80 c.c. 2.8 c.c. to 3.9. c.c. 0.36 c.c. to 0.40 c.c. For 1 per cent. Cinchonine sul- phate 3.1 c.c. to 4.0 c.c. 1.8 c.c. to 2.0 c.c. Kerner's test, in whether the article t the pharmacopoaial ested does or does form, merely determines not reach a certain recog- nized limit of impurity ; but, as applied by the analyst, the am- monia-water should be added from the burette and the required number of c.c. should be noted, as an index of the degree of impurity, whether above or below the legal standard. The number of c.c. of ammonia-water (of pharmacoposial strength and under pharmacopoeial conditions) is in itself a certain measure of value, already having a meaning to dealers and consumers, irrespective of interpretations in per cent, of cinchonidine or 1 Experiments by Mr. E. A. RUDDIMAN, made in an investigation now in progress in the University of Michigan, Indicate that of ammonia water of s. g. 0.960 there are required only 1.5 times more than of the water of s. g. 0.920, though the latter is 2.2 times stronger than the former. Averages of ten titra- tions, for each degree between 15 and 25 C., agreed nearly with this ratio of 1.5 to 1.0. * In respect to cinchonine (in presence of much quinine) these data are sur- prising. Taken separately, each in a cold-saturated sulphate solution (15 C.), Kerner in 1862 found the quantities of ammonia used to redissolve the alkaloid from 1 c.c. of the filtrate as follows: Of ammonia- water of s.g. 0.960, for qui- nine, 1.3 c.c. ; for cinchonidine, 16 c.c. ; for quinidine (b Quinidine). 15 c.c. ; for cinchonine, over 300 c.c. The experiments made by Mr. TEETER (Univ. Mich., 1880: New Rem., g, 258) in defining the limits of the test of quinine sulphate only show that in the preponderating presence of quinine both quinidine and ciuchonine require more ammonia than cinchonidine does. H2 CINCHONA ALKALOIDS. other alkaloids. The ammonia measure, based upon fixed condL tions of application, may be adopted over the world as a simple expression of comparative value. Against this preference for the ammonia measure it can hardly be urged that there is dis- agreement as to how many per cent, of cinchonidine are ad- mitted under 7 c.c. or 6 c.c. of ammonia. There may be dis- agreement as to the interpretation of any method of valuation. Kernels volumetric estimation of cinchonidine in commer- cial quinine sulphate is only an elaboration of his limit-test, so devised that the result is verified by a control analysis in each operation. It is as follows : * Standard Quinine Sulphate (Normalchininsulphat) is pre- pared by recrystallizing the salt from hot solution, with such a slight addition of sulphuric acid as shall give a faint acid reac- tion, usually crystallizing from three to six times, and until two portions of a crop of crystals, macerated at about 15 C., the one in 10 parts of water and the other in 500 to 700 parts of water, in parallel conditions, on titration of 10 c.c. of filtrate with am- monia-water, require the same number of c.c. of ammonia for solution. Usually by the recrystallizations the salt becomes neutral in reaction. The Standard Quinine Sulphate Solution is prepared freshly for use by rubbing the salt with about 100 times its weight of water in a mortar, rinsing into a glass-stop pered bottle, 2 and digesting along with the commercial quinine sulphate to be estimated, in the same conditions of temperature and time, as directed below. Water of Ammonia of sp. gr. 0.920 of ordinary quality is all that is required as a reagent. In the titration 5 grams of the sulphate of quinine to be tested are rubbed in a mortar with distilled water enough, so that when all is rinsed into a glass-stoppered bottle it shall just reach a mark of 50 c.c. volume. This bottle and the bottle containing the standard quinine sulphate solution are now set in the same ves- sel of cold water, at as near 15 C. as convenient, and left, with occasional careful shaking, for 12 to 18 hours. Or both bottles are warmed in the same vessel of water at near 100 C. for some time, shaking several times, and then set together in a vessel of cold water for an hour or more. The bottle containing the am- monia is placed in the same cold water, so that at the end of the J KERNEE, 1862. Improved in 1880: Archiv d. Phar., [3], 16, 186-285; 17, 438-454; Jour. Chem. Soc., 40, 63; New Rem., 10, 168. 2 The standard quinine solution should be strictly neutral in reaction. If acidulous, it is to be brought back to the neutral point by adding to it, with agitation, just sufficient of quinine hydrate, freshly precipitated from the same solution and well washed. A. B. P. QUININE. 143 digestion it shall have the same temperature. The two quinine solutions are now filtered through two dry filters, 1 at the ordi- nary atmospheric temperature (which is preferably near that of the digestion), obtaining from the standard quinine solution the same volume of filtrate furnished by the other solution (40 c.c. or over). 10 c.c. of each of these solutions is taken, by a good pipette, in a test-tube for titration. The ammonia is added from a burette, which is better if it be long, and narrow enough to register in -fa c.c. At first 5 c.c. of the ammonia are run in, the test-tube closed by the finger and given two or three circular motions to mix the liquid without shaking, and further smaller additions made, 0.3, 0.2, 0.1 c.c., and by drops, with the circular agitation after each addition, until the liquid becomes perfectly clear. Toward the last it is well to wait 5 to 10 seconds after each agitation before the next addition. The end reaction is complete clearing. Then at once the standard quinine solution is titrated in the same way, taking a fresh portion of ammonia in the burette. The 40 c.c. will suffice for four titrations of each quinine solution, from which the average can be taken. Each 0.32 c.c. (or 0.3, round number, the extremes being 0.28 and 0.35 c.c.) of the excess of the ammonia required for the qui- nine under test (beyond that required for the standard quinine) indicates 0.001 gram of cinchonidine sulphate. This 0.001 gram of cinchonidine sulphate is estimated upon the commercial qui- nine salt represented by the (10 c.c.) portion of filtrate taken,, or (having taken 10 c.c. of a 1 : 10 solution) each 0.32 c.c. of 0.920 ammonia (beyond that taken for the standard sulphate) indi- cates 0.1 per cent, of the cinchonidine impurity. Should the percentage of cinchonidine be over 1. 5, or at most 2.0, the results become inaccurate, owing to the gelatinizing of the precipitated alkaloid. In this case 10 c.c. of the filtrate under estimation may be diluted, by addition of standard quinine filtrate (of parallel digestion), to 20, 30, or 40 c.c., and portions of 10 c.c. of this diluted filtrate titrated. Then, after deduct- ing the average c.c. of ammonia taken by 10 c.c. of standard quinine, the remaining c.c. are multiplied by 2, or 3, or 4, when each 0.32 c.c. = 0.1$ cinchonidine, as before. The errors are stated not to exceed 0.05 per cent, of the commercial quinine salt. J The crystalline residues in the filters are to be saved, as purified sulphate of quinine, drying them by pressing the filters between blotting-papers, etc. It will be observed that the residue from filtration of the "standard quinine sulphate" solution is " standard quinine sulphate" prepared with an addi- tional purification. c44 CINCHONA ALKALOIDS. A.n approximate volumetric estimation is made in a short operation, according to Kerner (where last quoted), as follows : The quinine salt to be tested is macerated with ten times its weight of water at 15 0., 5 c.c. of the filtrate is taken in a test- glass of 10 c.c. capacity graduated in 0.1 c.c., 3 c.c. of water of ammonia of sp. gr. 0.920 are added and intermixed by gentle circular agitation of the test-glass while covered by the linger, and additions further made, at last by drops, until a clear liquid is attained, when the total volume is read and the volume of added ammonia is. noted. The required addition of only 3.0 to 3.3 c.c. of the ammonia- water would indicate absence of cincho- nidine sulphate; use of 5 c.c. ammonia (0.920) indicates near 1 per cent, cinchonidine sulphate ; and these data serve to show approximately the indication * (Kerner). The U. 8. Ph. (1880) directions for Kernels test are as fol- lows : '" The residue of 1 gram of (crystallized) sulphate of qui- nine, dried to a constant weight at 100 C. for estimation of water, is agitated with 10 c.c. of distilled water, the mixture macerated at 15 C. (59 F.) for half an hour, then filtered through a small filter, 5 c.c. of the filtrate taken in a test-tube, and 7 c.c. of water of ammonia (sp. gr. 0.960) then added; on closing the test-tube with the finger, and gently turning it until the ammonia is fully intermixed, a clear liquid should bo obtained. If the temperature of maceration has been 16 C. (60.8 F.), 7.5 c.c. of the water of ammonia may be added ; if 17 C. (62. 6 F.), 8 c.c. a may be added. In each instance a clear liquid indicates the ab- sence of more than about 1 per cent, of cinchonidine 3 or quiiii- dine, and of more than traces of cinchonine." The Ph Germ. (1882) directions are these : 2 grams of quinine sulphate are agitated with 20 c.c. of water at 15 C., and after half an hour filtered. To 5 c.c. in a test-tube am- monia [0.960] is added until the precipitated quinine is again dissolved. The required quantity of ammonia should not overgo 7 c.c. The Ph. Fran. (1884) directs the 2 grams quinine sulphate 1 By a more minute calculation, if the ammonia- water hold its strength, each 0.32 c.c. added above about 3.3 c.c. indicates 0.2 percent, of cinchonidine; so that 1 per cent, of cinchonidine is indicated by 4.9 c.c. (total addition), and 2 per cent, by 6.5 c.c. But exactness is not to be assumed without the help of the control analysis. A. B. P. Differences "of 0.5 C., Kerner states, do not sensibly affect the result. 2 These official allowances for temperature are more liberal than Kerner's results would justify. ^See the table from Kerner's figures, p. 141, according to which (at 15 C.) from 7.0 to 7.5 c.c. are required for 1 per cent, of cinchonidine sulphate. QUININE. 145 in 20 c.c. of water to be digested hot for half an hour, then maintained at 15 C. by immersion in a bath of water of this temperature for half an hour with frequent agitation, and filtered. Of the filtrate, 5 c.c. are treated with 7 c.c. of ammonia-water of 0.960 sp. gr., when a precipitate after gentle intermixture, or a turbidity or crystalline deposit formed after 24 hours, indicates an unacceptable proportion of alkaloids other than quinine. Another portion of 5 c.c. is evaporated in a tared capsule to a weight constant at 100 C., when the weight of the residue should not exceed 0.015 gram. 1 Temperature of the Filtrate in the reaction of the ammonia. Hitherto the influence of temperature has been regarded only as affecting the solubility of the sulphate of quinine, and the concentration of this salt in the filtrate. The temperature of digestion has been regulated, while that of the filtrate and the ammonia- water has been left to vary with the warmth of the atmosphere. From experiments recently made by Mr. E. A. Kuddiman, in the laboratory in which the author is engaged, it appears that the temperature of the filtrate under addition of the ammonia is influential. With digestion and filtration at 15 C., the warmer the filtrate becomes, the less ammonia is required to redissolve the quinine. For each 1 C. increase of temperature in the titration, an average of 0.148 c.c. less of ammonia of sp. gr. 0.920 is required to redissolve the quinine. This average was drawn from over ten titrations for each degree between 14 C. and 26 C., the temperature of the filtrate being taken at the end of the titration, the filtration itself being always held with the digestion at 15 C. The extremes were 0.1 and 0.2 c.c., for 1 C. 5n volumetric estimation by comparison with standard quinine sulphate, this influence of temperature of titration of the quinine will be the same in each of the parallel operations, and there- fore will not vitiate the conclusion. But in the pharmacopoeial tests, differences of titration temperature must affect the result, in part counteracting like differences of temperature in the digestion. The effects of titraticn temperature upon the cin- chonidine, cinchoriine, and quinidine are questions under in- vestigation. In 1884 Mr. HENRY B. PARSONS 2 reported the application of the U. S. Ph. form of the test to 1033 samples of quinine sul- 1 The residue of pure quinine sulphate would be 0.00675, leaving 0.00825 to consist of other alkaloids or impurities a quantity constituting about 1.6 per cent, of the commercial quinine sulphate tested. 5 " The Practicability of Kerner's Test ": Proc. Am. Phar., 32, 458. 146 CINCHONA ALKALOIDS. phate, embracing 5 brands, of American, German, and Italian production, as follows : Brand. No. samples. Average c.c. Am. Over 1 c.c. Am. No. 1. American. 16 9.5 16 " . 2. " 217 5,7 1 " 3. German. 11 6.1 none " 4. " 627 6.0 7 " 5. Italian. 162 6.8 35 Total, 1033 6.1 c.c. av. 59 (rejected). Mr. Parsons states that i ' if the sample of quinine sulphate be dried before testing, as the U. S. Ph. directs, the amount of am- monia-water required to produce a clear solution is generally, but not always, about 0.5 c.c. greater than where the same sample is not dried before testing." Also, "the test is liable to mislead unless every detailed precaution is observed." In 1884 B. F. DAVENPORT, as State Analyst of Drugs in Massachusetts, 1 ex~ amined 28 samples, from seven makers, using the official form of Kerner's test, and found -28 per cent, of the samples to fall below the U. S. Ph. requirement. The application of the Ammonia Test to Quinine Com- pounds other than the Sulphate requires their conversion into sulphate. 2 This may be done, in an exact application of the test to salts of quinine* other than sulphates, very easily as follows : A weighed quantity, ^rom 2 to 5 grams, of the salt is dissolved in about fifty times iid weight or a sufficient quantity of water, the alkaloids completely precipitated with sodium hydrate solu- tion, the precipitate washed until the washings give but little cloudiness with magnesium salt solution, and the washed preci- pitate rinsed through a perforation in the filter-point into a test- glass, graduated in \ c.c. and measuring 20 to 50 c.c., filling to near the volume specified below. The mixture is heated for five minutes by immersing the test-glass in nearly boiling water, 1 " Fifth Ann. Report Mass. State Board of Health," etc., Boston, 1884, p. 161. 2 In his first paper, in 1862, Kerner proposed to apply the ammonia test directly to salts not sulphate, directing the solution of the salt to be diluted to the limit of solubility of quinine sulphate (Zeitsch. anal. Chem., I, 161). The later report (1880) does not reach the application to other salts. 3 To find, for any salt of quinine, the volume equal to 10 c.c. for each gram of crystallized normal sulphate producible from 1 gram of said salt: Comb. no. of salt taken (in equation to form. 1 mol. sulphate) : 872 : : 10 : x = c.c. de- sired. QUININE. 14; and dilute sulphuric acid is added to maintain a slight acid re- action to litmus-paper during the digestion. The mixture is now exactly neutralized to litmus-paper by adding dilute' am- monia-water, the volume of the whole made up to a number of c.c. equal to 11.5 times the number of grams of quinine hydrochloride taken, 10.3 " " " hydrobromide " 9.8 " " " valerianate " when the mixture is placed for half an hour or longer in a bucket of water at 15 C. (59 F.), and filtered through a small filter. One or more portions of 5 c.c. are tested with ammonia-water, as in the pharmacopoeial form of the test (see page 144), and the result judged for the salt taken, on the basis of quinine sulphate. Or, cooling parallel with " standard quinine sulphate " solution, for titration, as directed on p. 142, portions of 10 c.c. are titrated in comparison with " standard quinine," for percentage of cincho- riidine, etc. The results will count, on the basis of the 5 c.c. or of the 10 c.c. of filtrate used, in per cent, of the salt of quinine taken. That is, each 0.32 c.c. of ammonia of sp. gr. 0.920 used for 10 c.c. of filtrate (beyond that used for the "standard quinine'') indicates 001 gram, or 0.1 per cent., of cinchonidine hydrochlo- ride in the commercial quinine hydrochloride taken, or of cin- chonidine hydrobromide in commercial quinine hydrobromide taken, etc. Under Hydrochlorate of Quinine the Br. Ph., 1885, states that " it may be converted into sulphate of quinine by dissolving it, together with an equal weight of sulphate of sodium, in ten times its weight of hot distilled water, and setting the mixture aside at 60 F. (15.5 C.) for half an hour. Such sulphate should respond to the characters and tests," etc., no further directions being given. The Ph Fran., 1884, does not apply tests for cin- chonidine to hydrochloride or hydrobromide of quinine. The Ph. Germ., 1882, directs to evaporate 2 grams of hydrochloride of quinine with 1 gram of sodium sulphate and 20 grams of water, to dryness, digest the residue with 12 grams of alcohol, evaporate the filtrate, and subject the resulting quinine sul- phate to the test prescribed for this salt. The U. S. Ph., 1880, directs for quinine hydrochloride and hydrobromide alike that " 1.5 gram be dissolved in 15 c.c. of hot distilled water, the solu- tion stirred with 0.75 gram [for the hydrobromide, 0.60 gram] of crystallized sulphate of sodium in powder, the mixture main- tained at 15 C. for half an hour, and then drained through a filter only large enough to contain it, until 5 c.c. of filtrate are 148 CINCHONA ALKALOIDS. obtained ; upon treating this liquid as directed for the corre- sponding test under quinine (p. 144) the results there given should be obtained." F. B. POWER* states that, following the U. S. Ph. directions, he obtained only from 2 c.c. of filtrate with the hydrobromide, and only about 1 c.c. of filtrate in testing the hydrochloride ; and he proposes taking 30 c.c. of water instead of 15 c.c. for the 1.5 'grams of hydrobromide or hydrochloride of quinine. C. N. LAKE" has avoided the difficulty in another way, adopted in a habitual use of the test upon these salts, namely : by repeatedly adding water and evaporating to dry ness, with stirring, whereby the bulkiness of the precipitated mass be- comes reduced, and the 5 c.c. of filtrate are obtained. The ob- vious remedy for deficiency of the filtrate (as remarked also by Mr. Lake) is to take larger quantities of the materials without altering their proportion to the water (see p. 139). Certainly the stated proportion of water is an influential factor in the test, not to be varied unless a correction be needful to preserve the conventional ratio 3 of 1 to 10 between the sulphate and the water. Such a correction, as seen on p. 147, would make a slight but appreciable diiference with the hydrochloride, not an appreciable difference with the hydrobromide. Greater diffe- rences are probably due to the presence of sodium sulphate, and bromide or chloride, in the U. S. Ph. application of the test. At all events, the taking of twofold or threefold the quantities of the salts and the water directed by the U. S. Ph., in its specified proportions, does not constitute a departure from its authority for these tests. In the ammonia test for purity of Quinine, free alkaloid, a weighed quantity, from 2 to 5 grams, either of the hydrate or of the anhydrous alkaloid obtained by drying to a constant weight at 100 115 C., may be converted to normal sulphate by digest- ing with warm diluted sulphuric acid, as directed for the pre- '1885: "Contributions Dept. Phar. Univ. Wis.," p. 11. 2 1885: "The Ph. Application of Kerner's Test to Quinine and its Salts ": Drug. Cir, 29, 199 (Oct.) 3 It is the concentration of the solution of alkaloids not quinine that is to be guarded against variation in the test a test for the quantity of these alkaloids by measure of the concentration of the ammonia needful to dissolve them. The quinine sulphate concentration is securely constant, in solution sure to be saturated. The concentration of the solution of "other alkaloids" is not un- derstood to be affected by the absorption of a good part of this solution by a bibulous mass of imperfectly crystallized salt. If the proportion of water were materially varied by entering into combination as crystallization-water, this variation would be a proper subject of correction. But in the two cases in hand more crystallization- water is liberated than is taken up, the difference being immaterial. QUININE. 149 cipitated alkaloid on p. 146, making up to the conventional volume of the strictly neutral mixture and treating further, as there directed. Grams of quinine hydrate taken X 11.5 (or grams of anhydrous quinine X 13.4) = c.c. of conventional volume. Each 0.32 c.c. of ammonia-water of sp.gr. 0.920 used intitration of 10 c.c. of filtrate indicates 0.001 gram, or 0.1 per cent., of free cinchonidine in the commercial free quinine tested, etc. For 5 c.c. of filtrate not over 7 c.c. of ammonia-water of sp. gr. 0.960 should be required, to correspond to the pharmacopoeial standard for quinine sulphate. The IT. S. Ph. test for Quinine (hydrate) converts the alka- loid into the sulphate by drying on the water- bath a wetted mixture of the quinine with half its weight of ammonium sulphate. 1 In the ammonia test for purity of Quinine Bisulpliate, it may be converted to the normal sulphate, without loading the so- lution with alkali sulphates, as follows : A weighed quantity, from 2 to 5 grams, of the bisulphate is dissolved, in a graduated test-glass, in about 12 times its weight of warm distilled water. The solution is carefully divided into two exactly equal portions by volume; from the one portion (taken in a small beaker) the alkaloid is fully precipitated by sodium hydrate solution, and the precipitate washed on a filter until the washings are made but slightly cloudy by barium chloride solution, when the drained precipitate is partly dried by blotting-paper and transferred from the filter to the other portion of the bisulphate solution, in the graduated test-glass (p. 146). The mixture is now heated (by immersing the test-glass in hot water), exactly neutralized to litmus-paper by adding dilute sulphuric acid or ammonia, and the volume made up to a number of c.c. equal to 8 times the number of grams of the bisulphate first taken, when the mix- ture is crystallized at 15 C. and further treated as directed on p. 144 or 142, taking 5 c.c. of the filtrate for the limit-test, or 10 c.c. for titration parallel with " standard quinine solution " to estimate cinchonidine. For 5 c.c. of the filtrate not over 7 c.c. of ammonia- water of sp. gr. 0.960 should be used, to correspond 1 Prof. Power (where cited on p. 148) found that with 1 gram quinine taken only 3 c.c. filtrate could be obtained. Mr. Lake (where cited, p. 148), taking the 1 gram quinine, obtained the required 5 c.c. of filtrate, in his practice as an analyst, by evaporating to dryness and adding water, which, he says, it was not necessary to do over three times. To take 2 to 5 grams of the alkaloid, with half its weight of ammonium salt, and the tenfold number of c.c. of water, in- volves no departure from the authority of the pharmacopoeia for the require- ment. ISO CINCHONA ALKALOIDS. with the pharmacopceial standard for normal quinine sulphate. And for 10 c.c. of the h'ltrate each 0.32 c.c. of ammonia-water of sp. gr. 0.920 used (beyond that used for the " standard quinine") indicates 0.1 per cent, of cinchonidine bisulphate (cryst., 2H 2 O, HESSE) in the article under examination. The U. S. Ph. directs that the dried salt be neutralized with ammonia, made up to 10 fluid parts for 1 part of crystals taken, then held at 15 C. and further treated as in the test of the nor- mal sulphate. The same simple operation (drying the salt with the ammonia) is directed by the Ph. Germ. The presence of the ammonium sulphate resulting from the neutralizing with ammonia probably adds severity to the test ; while the dilution to 10 fluid parts for 1 part of bisulphate amounts to 12.5 fluid parts for 1 part of normal sulphate formed, and certainly dimin- ishes the severity of the test. The U. S. Ph. makes the opera- tion upon 1 gram of the salt, and the Ph. Germ, upon 2 grams of the salt. POWER 1 has proposed to take 1 gram of the salt with 20 c.c. of water, because 10 c.c. of water fail to yield 5 c.c. of filtrate. Where the operator is unable to depart from the pharmacopoeia, he should preserve the proportions of 1 to 10 for the grams of crystallized salt to the c.c. of the mixture digested at 15 C. If any departure be made in this proportion, it should be the adoption of the ratio of 1 to 8, when the salt is neutralized with ammonia or neutralized with its own alkaloid obtained from a divided portion of that taken. The ammonia test of Effloresced Salts of Quinine. The Sul- phate and the Bisulphate are frequently effloresced when taken for the ammonia test, and the hydrate is apt to have less than 3H 2 O. The severity of the test is thereby increased, in propor- tion to the resulting increase of concentration of the alkaloids not quinine. All deviations of the concentration may be avoided by drying the salts to a constant composition, taking the anhy- drous or the effloresced form by weight, and making up the volume for digestion at 15 C. according to the following data. Then 5 c.c. (or 10 c.c.) of the filtrate will in each instance have the same conventional limit of concentration for the sulphates of all the cinchona alkaloids : 1 Where cited on p. 148. The recommendation is to increase the propor- tion of water, when it is already 25 percent, too large. The pharmacopoeia does not take 1 gram of the dried salt, as Prof. Power's equation (p. 14, loc. cit.} represents, but " 1 gram of the salt " the words " previously dried at 100 C." qualifying "agitated with 8 c.c. of distilled water," and the procedure of dry- ing being parallel to that more explicitly laid down for quininae sulphas. LAKE (where cited on p. 148) has used the 1 gram with 10 c.c., and obtained the need- ful 5 c.c. of nitrate by evaporating to dryness, stirring, and adding water. QUININE. Taking 1 part by weight or " " grams. Weight constant, C. Molecules water. Per ct. crystal, water. Fluid parts (tinges " n "c.c.) of mixture. Crystallized quinine sulphate. 7 14 45 10 Effloresced " " See p. 126. 2 4.60 11 2 Anhydrous " " 100 115 11 7 Crystallized quinine bisulphate 7 23 00 8 O 1 Effloresced " .... Anhydrous " " .... Quinine hydrate . . See p. 127. 100 See p. 126 1 8 4.09 14 28 10.0 10.3 11 5 Anhydrous quinine 100 120 13 4 IS 'erze- Kindt Hesse's test for Quinine Sulphate (see p 116) differs in prin- ciple from Kerner's in using ether, instead of excess of ammonia, as a solvent of the quinine, and differs from Liebig's test* in 1 If the bisulphate be neutralized with its own alkaloid obtained from a divided portion of the salt weighed and taken, the ratio is the same. 2 "Liebig's Test." The author is unable to cite published directions of Liebig for the test which has gone under his name for fully thirty years. The test, in a simple form, with far too large proportions of ether and ammonia, *~ very clearly given, in 1833, on the sole authorship of KINDT, as follows (Ber. Urn's Jahresbericht, 12 (1833), 218; from Branded Archiv, 36, 254): "Kin hat folgende Methode zur Entdekkung der Gegenwart von Schwefelsaurem Cinchonin im Chininsalz angegeben. Man zerreibt 1 Gran vom Salze, schiittet es in em Probirglas, und giesst 1 Drachme Aether darauf, womit man es um- schilttelt; alsdann mischt man 1 Drachme Ammoniak zu und schiittelt wolil um. Wenn sich die Fliissigkeiten wieder scheiden, findet man die Scheidungs- linie rein, wenn das salz frei von Cinchonin war, aber die geringste Menge Cin- chonin im Salz setzsich deutlich erkennbar aus der Grenze zwischen beiden Fliissigkeiten ab." In 1842 CALVERT (Jour, de Phar ; Ann. Chem. Phar., 48, 242) undertakes to separate cinchonine by its insolubility in calcium chloride solution and in lime solution, and refers to the use of ammonia, but not to the use of ether, as a solvent of quinine. In 1843 R. HOWARD (Phar. Jour. Trans., 2, 645) says that cinchonine sulphate is not at all excluded from commercial quinine sulphate ' by any test " which he has " happened to see recommended," and he gives a test by weight of crystals from a saturated sulphate solution. In 1852 SOUBEIRAN (Jour, de Phar., 1852, Jan.; Am Jour. Phar., 24, 166) cites Liebig's authority for the ether test, saying only "Liebig has suggested" the detection of cinchonine by treating 15 grains of the salt, with 2 ounces ammonia solution, and 2 ounces of ether, and so on, the proportions being nearly those given by Kindt in 1833, though the quantities are 15 times as large. From about this time (1852) the test is commonly menlimed in literature as Liebig's test; but in 1851 or 1852 ZIMMER (Jahr. Chem., 1852, 745; Chem. Vazette, 1852, 449) gives the test, with the modern proportions of ether and of ammonia, with- out naming Liebig's authority. Also HENRI (1847) separates cinchonine by ether in a complex process, which is criticised by GUIBOURT in 1851-52, neither of these authors speaking of Liebig. In Gmelin's Chemistry (Cav.ed. 17, 279) the method is entitled "The Quinine Test of Liebig," but among the references to as many as a dozen published authorities Liebig's name is not found. In this historical inquiry it may further be noted that when LIEBIG reported the elementary analyses of quinine and cinchonine, in 1831 (Ann. Phys. them., 152 CINCHONA ALKALOIDS. removing the excess of the quinine as a sulphate before separat- ing by .ether. The directions of Hesse are as follows : 1 A " qui- nometer " is provided, this being a test-tube 10-11 mm. (0.39 to 0.43 inch) wide, and 120 mm. (4.7 inches) long. The tube is marked at a capacity of 5 c.c., and again at a capacity of 6 c.c. ; the entire capacity of the tube, which is fitted with a cork, being 10 or 12 c.c. Of the quinine sulphate to be tested, 0.5 gram is well shaken in a test-tube with 10 c.c. of hot water (50 to 60 C.), and set aside to cool for ten minutes, shaking with care to prevent the expulsion of the contents. The liquid is now passed through a filter of about 60 mm. (2.4 inches) diameter into the quinome- ter, up to the 5 c.c. mark ; 1 c.c. of ether (sp. gr. 0.724 to 0.728) is added (up to the 6 c.c. mark), and then 5 drops of ammonia- water (sp. gr. 0.96), when the tube is corked and slowly shaken. Gra- nular crystals appearing within 3 minutes after shaking indicate as much as 3 per cent, of cinchonidine ; at 10 minutes after shak- ing, about 2 per cent, of cinchonidine. After standing 2 hours the appearance under a lens of granular crystals indicates cincho- nidine ; radiating needles, cinchonine or quinidine ; no crystals, the absence of over 1 per cent, of cinchonidine, or 0.5 per cent, of quinidine, and of over 0.25 per cent, of cinchonine. Absence of crystals after 12 hours shows that less than 1 per cent, of cin- chonidine is present. If now the cork be loosened, and the ether permitted slowly to evaporate, 0.5 per cent, of cinchonidine will leave a distinct crystalline residue. The final residue con- tains amorphous quinine. The test of Sulphate of Quinine for Cinchonidine by the Br. Ph., 1885, on the principle of Hesse's test, is much more elaborate than the operation above detailed : " Test for Cincho- Pogg., [3], 21, 25), he specifies the purification of quinine by dissolving it, not in ether, but in ammonia (on the plan of Kerner's test), *as follows: "Der breiartige weisse Niederschlag, welcher durch verdunstes Armnoniak aus der schwefelsauren Auflosung erhalten worden war, loste sich beim Erhitzen in der etwas freies Ammoniak enthaltenden Fliissigkeit vollkomrnen auf, und gab bei dem abkiihlen ganz Ammoniak freie, sehr feine, glanzende, seidenartige Nadeln von Chinm." Liebig's test was directed by the U. S. Ph. of 1860 and of 1870, in the fol- lowing terms: "When 10 grains of the salt [Sulphate of Quinine] are agitated in a test-tube with 10 minims of officinal water of ammonia [0.960] and 60 grains of ether [0.750], and allowed to rest, the liquid separates into two trans- parent and colorless layers, without any white or crystalline matter at the sur- face of contact." It was generally stated that as much as 10 per cent, of quini- dine sulphate would escape detection by this test. Undoubtedly larger percen- tages both of quinidine and cinchonidine are liable to fail of recognition with the test as commonly applied. 1 O. HESSE, 1878': Archiv d. Phar., [3]. 13, 490; Am. Jour. Phar., 51, 135; New Rem., 8, 139; Jour. Chem. Soc., 36, 280; Zeitsch. anal. Chem., 19, 247. QUININE. 153 nidine and Cinchonine. Heat 100 grains (6.48 grams) of the sulphate of quinine in five or six ounces (142-170 c.c.) of boiling water, with three or four drops of diluted sulphuric acid. Set the solution aside until cold. Separate, by filtration, the purified sulphate of quinine which has crystallized out. To the filtrate, which should nearly fill a bottle or flask, add ether, shaking oc- casionally, until a distinct layer of ether remains undissolved. Add ammonia in very slight excess, and shake thoroughly, so that the quinine at first precipitated shall be redissolved. Set aside for some hours or during a night. Remove the superna- tent clear ethereal fluid, which should occupy the neck of the vessel, by a pipette. "Wash the residual aqueous fluid and any separated crystals of alkaloid with a very little more ether, once or twice. Collect the separated alkaloid on a tared filter, wash it with a little ether, dry at 212 F., and weigh. Four parts of such alkaloid correspond to five parts of crystallized sulphate of cinchonidine o,r of sulphate of cinchonme. " Test for Quinidine. Recrystallize 50 grains (3. 240 grams) of the original sulphate of quinine as described in the previous paragraph. To the filtrate add solution of iodide of potassium, and a little spirit of wine [alcohol] to. prevent the precipitation of amorphous hydriodides. Collect any separated hydriodide of quin- idine, wash with a little water, dry and weigh. The weight repre- sents about an equal weight of crystallized sulphate of quinidine. " Test for Cupreine [see p. 92]. Shake the recrystallized sulphate of quinine, obtained in testing the original sulphate of quinine for cinchonidine and cinchonine, with one fluid-ounce (28.4 c.c.) of ether (sp. gr. 0. 724-0. 728) and a quarter of an ounce (7.1 c.c.) of solution of ammonia (of 10$ strength), and to this ethereal solution, separated, add the ethereal fluid and wash- ings also obtained in testing the original sulphate for the two alkaloids just mentioned. Shake this ethereal liquor with a quar- ter of a fluid-ounce (7.1 c.c.) of a ten per cent, solution of caustic soda, adding water if any solid matter separates. Remove the ethereal solution. Wash the aqueous solution with more ether, and remove the ethereal washings. Add diluted sulphuric acid to the aqueous fluid heated to boiling, until the soda is exactly neutralized. When cold collect any sulphate of cnpreine that has crystallized out, on a tared filter, dry, and weigh. " ' Sulphate of Quinine ' should not contain much more than five per cent, of sulphates of other cinchona alkaloids." Water of Crystallization in Sulphate of Quinine. HESSE 1 '1880: Ber. d. chem. Oes., 13, 1517-1520, and elsewhere. 154 CINCHONA ALKALOIDS. has continued to maintain that perfect crystals of pure quinine sulphate have 8H 2 O (16.18$ of water). KEENER' affirms that no quinine sulphate is manufactured that contains over 14 to an extreme of 14. 6$ of crystallization-water ; above this any con- tained water is free moisture. Also that it is not possible to dry the voluminous quinine sulphate of commerce without some degree of efflorescence. The Ph. Germ. (1882) (without for- mulae) limits the loss by drying at 100 C. to 15$. The Ph. Fran. (1884) gives YH 2 O (=14.45$) in the formula, and limits the loss at 100 C. to 14.45$. The Br. Ph. (1885) gives 7jH 2 O in the formula, and limits the loss at 100 C. to this molecular proportion, 14.3$. The U. S. Ph. (1880) gives 7H 2 O in the for- mula, and limits the loss of weight at 100 C. to 16.18$. Mr. H. B. PAKSONS 2 reported the loss of water by drying three hours in a (boiling) water-oven, for 1015 samples, of American, Ger- man, and Italian makers, each sample representing 100 ounces, and taken from a can not previously opened. T^he average of loss of water, for all the samples, was 13.84$ ; for any single manufacturer the lowest average was 12.61$, and the highest average was 14.36$. The samples of one maker all approached closely to 12 53$ (6H 2 O). In the report the writer recommends, as others have done, the pharmacopoeial adoption of effloresced quinine sulphate, the two-molecule salt, as a definite and stable form of the alkaloid. QUINIDINE. The Conchinine of Hesse 3 C 20 H 24 N 2 O 2 =324. In crystals with 2|H 2 O=i396. Chinidine. KationarFormula, p. 98 ; Proportion in Cinchona Barks, p. 97. Separation from the Bark, in total alkaloids, p. 102. Separation from Cinchona M880: Archiv d. Phar., [3], 17, 453. 2 1884: Proc. Am. Ptiarm., 32, 457. 3 The name quinidine, in German "chinidine," was given to the alkaloid now universally known as cinchonidine, in 1833, by HENRY and DELONDRE. Quinidine was itself discovered, in chinoidine, in 1849, by VAN HEIJNINGEN, who then named it ^-quinine; again, in commercial cinchonine, in 18.il, by HLASIWETZ, who named itcinchotine. In 1853 PASTEUR, believing that he iden- tified Henry and Delondre's quinidine among cinchona alkaloids, and discover- ing another which in fact was Henry and Delondre's quinidine, he fixed to this the name cinchonidine, still retained. The name "quinidine" having thus been differently applied, HESSE (1874) proposes to drop it, and use the name " conchinine" for the isomer of quinine. Some of the German writers employ Hesse's nomenclature, but English-writing chemists translate the "conchinine " of Hesse into the English equivalent of German ' chinidine," namely: quini- dine. And this accords with the recommendation of the Quinological Con- gress at Amsterdam in 1877 Further see KERNER'S history of this nomen- clature, 1880: Archiv d. Phar., [3], 16 (reprints). The 6-quinidine of Kerner in 1862 is the quinidine of the present time. QUINIDINE. 155 Alkaloids, index at p. 112. Distinction from other Cinchona Alkaloids, index at p. 100. Microscopic identification, p. 101. .Rotatory Power, p. 123. The free alkaloid has been hardly known in commerce, and its sulphate is less used than that of cinchonidine or cinchonine. The chinoidine obtained as a by-product from certain barks is rich in quinidine. The crystalline forms and heat reactions of quinidine and its salts are given under a, the solubilities of the same under c, below. Quinidine is identified by its fluorescence in the sul- phate and its response to the thalleioquin test (d), together with the free solubility of the sulphate in chloroform and its greater solubility in water. Also by precipitation as hydriodide (d). It is separated, by solution of the sulphate in chloroform, or pre- cipitation with iodide, or otherwise (0); estimated, usually by weight of the hydriodide (/). Tests for impurities in quini- dine sulphate are presented under g, p. 157. a. Quinidine crystallizes from alcohol, with 2JH 2 O, in large, lustrous, monoclinic prisms or needles, efflorescent in the air. From ether permanent rhombohedrons with 2H 2 O are obtained ; from boiling water permanent plates with 1^H 2 O (HESSE). The whole of the water is removed at or below 120 C., and the dry al- kaloid melts at 168 C. It begins to brown very slightly at 160 C. (BLYTH). Quinidine sulphate, (C 2 pHo 4 N 2 O 2 ) 2 H 2 S6 4 .2H 2 O, crystallizes in white, silky needles or in long, hard prisms, per- manent in the air, giving up the water at 120* C. The hisulphate crystallizes in asbestos-like prisms, with 4H 2 O. Quinidine hy- drochloride crystallizes in asbestos-like fibres, with H 2 O. Qui- nidine oxalate, normal, crystallizes with H 2 O, in pearly plates or in prisms. b. In taste and physiological effects quinidine resembles quinine. c. Quinidine is soluble in 2000 parts of water at 15 C., in 750 parts of boiling water ; in 26 parts alcohol of 80$ at 20 C. ; in 22 parts of ether of sp. gr. 729 at 20 C., or in 35 parts of the same at 10 C. (HESSE, 1868). In 809 parts ether of 0.72 sp. gr. at 19 C. (VAN DEB BURG) ; in 76.4 parts of ether at 10 C. (DRAGENDORFF). In chloroform or amyl alcohol it is readily solu- ble; in petroleum ether difficultly soluble. Quinidine neutral- izes acids in forming normal salts. Quinidine sulphate of a neutral reaction is soluble " in 100 parts of water and in 8 parts of alcohol at 15 C. ; in 7 parts of 156 CINCHONA ALKALOIDS. boiling water, and very soluble in boiling alcohol ; also in acidu- lated water and in 20 parts of chloroform, but almost insoluble in ether " (U. S. Ph.) In 19.5 parts chloroform at 15 C., in 9 parts at 62 C. (HESSE, 1 879). Quinidine hydrochloride is solu- ble in 62.5 parts of water at 10 C.,-and freely soluble in hot water, in alcohol, and in chloroform ; nearly insoluble in ether. Quinidine hydrobromide, anhydrous (!)E VRIJ, 1875), is soluble in 200 parts of water at 14 C. Quinidine oxalate, (Co H 2 oN"oOp)oH 2 CoO 4 .HoO, dissolves in 150 parts of water at 15 C. d. In solutions of the sulphates, and especially in solutions acidulated with sulphuric acid, quinidine exhibits strong blue fluorescence. (See Quinine, d.) The chloroformic solution of the sulphate has a green fluorescence (HESSE, 1879). Quinidine responds to the thalleioquin test (p. 130). Sulphuric acid gives no color ; Froehde's reagent, a greenish color. Iodide of po- tassium causes in neutral solutions of quinidine salts a crystal- line precipitate of quinidine hydriodide, C 20 H 24 N 2 O 2 HI, soluble in 1250 parts of water at 15 C. (DE VRIJ). Immediate precipi- tation is obtained only in somewhat concentrated solution, and is incomplete. Full crystallization within the limit of solubility is obtained by warming the mixture and stirring it with a glass rod from time to time as it cools, then leaving some hours at a low temperature, stirring at intervals. The reagent should be neutral, and added in such proportion that the quantity of solid potassium iodide shall nearly equal the quantity of alkaloid in solution. The crystals slowly formed in dilute solutions are leaf -form. In acidulous mixtures of sufficient concentration bihydriodide of quinidine is formed, in golden crystals, soluble in 90 parts of water at 15 C. (DE YEIJ). With the alkalies and alkali carbo- nates quinidine gives nearly the same reactions as quinine, the precipitate being very much less soluble in excess of ammonia. In presence of quinine the quinidine precipitate requires a good excess of ammonia to dissolve it, and the precipitate is apt to re- appear, crystalline on standing. With the general reagents for alkaloids quinidine reacts nearly the same as quinine, so far as the reactions have been examined. The dextrorotatory power of quinidine is given on p. 123. e. Separations of quinidine are obtained chiefly (1) by its crystallization as hydriodide (d, /*), and (2), except from cincho- nine, by solution of the sulphate in chloroform (ERENS). It prevents the lactic, not the alcoholic, fermentation (DONATH). It is not found in the urine after administration. c. Soluble in water, sparingly when cold, freely when hot. Soluble in all proportions in alcohol, ether, and carbon disul- phide, and soluble in chloroform, benzene, amyl alcohol, carbon disulphide, and in fixed and volatile oils. The salts of quinoline are soluble in water. The tartrate, (C 7 H 9 N) 3 (C 4 H 6 O 6 ) 4 (FRIESE, BERNTHSEN, 1881), is soluble in 80 parts of water at 16 C.; in 150 parts of 90$ alcohol at 16 C. ; in 350 parts of ether. It melts at about 125 C. The hydrochloride, C 9 H 7 N.HC1 (OECHSNER, 1883), is soluble in water, alcohol, chloroform, ether, and benzene, in the last two solvents sparingly in the cold. Melts at 94 C., and vola- tilizes. d. Quinoline is indicated by its odor, obtained from its salts on addition of a fixed alkali. Alkali hydrates precipitate it, in solutions not dilute ; the precipitate being soluble in ex- cess of ammonia, and easily taken into solution by ether, chlo- roform, and other solvents of the base. Solutions of quinoline salts are precipitated by the general reagents for alkaloids. According to Donath, the limits of precipitation, in certain favo- rable proportions of reagents, were as follows : For iodine in iodide of potassium, 1 to 25000 parts ; phosphomolybdate, 1 to 25000 parts ; mercuric chloride, 1 to 5000 parts ; potassium mercuric iodide, 1 to 3500 parts. The precipitate with phosphomolybdate, yellow-white, dissolves colorless in ammonia ; with mercuric chloride, white. The precipitate by potassium mercuric iodide, 'BERENS, 1885: Ther. Gazette, 9, 433. , KAIRINES. 167 on adding hydrochloric acid, crystallizes in amber-colored needles. No color is caused by sulphuric or nitric acid (DONATH). By long heating with excess of sulphuric acid quinolirie sulphonic acid is formed. . Tests for impurities. In artificial quinoline by Skraup's process, nitrobenzene has been found as an impurity (0. EKIN, 1882). The salts should be completely soluble in water, the free base in water with sufficient acid. There should be no bitter taste (impurity from cinchonine). Alkali hydrates should not cause a colored precipitate. Cinchonine-quinoline, as prepared for use, and unless repeatedly distilled and recrystallized, con- tains lepidine (HOOGEWERFF and v. D ) ; and therefore when treated with amyl iodide, and then with caustic alkali, gives a blue color, formation of a cyanine (WILLIAMS), G^^CgEE-^. C 10 H 9 NC 5 H 11 I. Aqueous solution of pure quinoline salt [not alkaline] does not sensibly change the color of permanganate so- lution in the first eight or ten minutes (HAGER). KAIRINES. Methyl or ethyl substitutions in oxy-qumoline- tetrahydride, C 9 H 10 (OH)N. The methyl compound is C 9 H 9 (CH 3 XOH)N=:C 10 H: 13 NO ; the ethyl compound, C 9 H 9 (C 3 H 5 ) (OH)N=C 11 H 15 NO. The name kairine is used for the hydro- chloride. Oxyhydro-methylquinoline is termed Kairine M, and oxyhydro-ethylquinoline Kairine E or Kairoline. Derivatives of quinoline (E. FISCHER, 1883) of medicinal interest. The free bases are not stable in the air. &, c. The methyl base crystallizes in rhombic forms ; is spar- ingly soluble in water, soluble in alcohol, ether, and benzene, and acts as a strong base in forming salts. It boils at 114 C. The hydrochloride, C 10 H 13 NO . HCl-l-HsO, forms lustrous, monocli- iiic crystals, generally found in a slightly colored crystalline pow- der, easily soluble in water. At 110 C. it loses its water of crys- tallization and turns violet. The sulphate, (C 10 H 13 NO) 2 H 2 SO 4 , forms lustrous prisms. The ethyl base crystallizes in scales or plates, melting at 76 C., slightly soluble in water, freely soluble in alcohol, ether, and ben- zene ; hardly soluble in petroleum benzin. The hydrochloride, C 11 H 15 NO . HC1, forms white prisms, generally appearing in grayish-yellow crystalline powder, freely soluble in water, spar- ingly soluble in hydrochloric acid. b. The kairines have a bitter and saline, disagreeable taste and a penetrating odor. Ordinary doses are one-half to one gram 168 CINCHONA ALKALOIDS. to 15 grains), and doses of 25 to 50 grains cause disturbance. 1 t is in part excreted unchanged in the urine (MERING, 1884), The ethyl compound differs from the methyl compound only in a somewhat longer duration of effect (Filehne). d. Kairines are indicated by the penetrating, characteristic odor of the free base, obtained in full from the salts on adding a fixed alkali, and by the bitter taste. In aqueous or alcoholic so- lution, treated with oxidizing agents, as dichromate and an acid, they give rosaniline colors, violet-blue to violet-red, in some reactions greenish tints being obtained. Ferric chloride gives a brown color in solutions, with gradual precipitation. Sodium nitrite in sulphuric acid solution gives orange to red colors. Po- tassium ferrocyanide gives an abundant precipitate ; phospho- tungstic acid a pale yellow precipitate. When the base is libe- rated, as in alkaline solutions, the kairines rapidly oxidize in the air, with deposition of brown, humus- like bodies. THALLINE. C 10 H-,^N"O. Tetrahydroparaquinanisoil. A de- rivative of paraquinanisoil. a One of the methyl kairines, isomeric with " kairine M." Thalline appears in pale yellow crystals, melting at about 42 C., boiling at 282 C. without decomposition. Its salts are given in- doses of 0.25 to 0.75 gram. It is sparingly soluble in cold, more freely in hot water, and soluble in alcohol, ether, and petroleum ether. It makes stable salts ; but in all forms it is easy to suffer change, and the light affects it injuriously. The sulphate and tartrate are obtained in nearly white crystals or crys- talline powder, melting at 100 C., with browning. The sulphate is freely soluble in water, nearly insoluble in ether, but is some- what soluble in chloroform. Oxidizing agents produce an intense green color with thalline, hence its name. Ferric chloride is a favorable oxidizing agent for the purpose, giving a deep emerald- green color, not changed by acidulation with sulphuric acid, but changed by reducing agents. In physiological effect thalline resembles the kairines. 3 ANTIPYEINE. C n H 19 N 2 O. A proposed commercial name for Dimethyl-oxy-quinizine/ C 9 H 6 (N.CH 3 )(CH 3 )(O)N, the hypo- thetical base quinizine having the general formula C 9 H 9 (NH)]Sr 'On use of kairine as an antipyretic, FILEHNE, 1883-1883. American uses summarized in Ther Gazette, 9, 122 (Feb.. 1885). 2 VuLPius, 1883: Archiv d. Phar., [31, 22, 840; Jour. Chem. Soc., 1885, Abs.,398, 1022. 3 BEYER, 1886: Am. Jour. Phar., 58, 196. JAKSCH, 1884 ANTIPYRINE. 169 (L. KNOKR, 1884 1 ). Of interest for medicinal uses as an anti- pyretic. a. Antipyrine crystallizes in needles, melting at 113 C. In commerce it appears as a white, crystalline powder, sometimes slightly colored. b. Of a very mild bitter taste, not disagreeable, and a barely perceptible odor. Dose, 1 to 2 grams (15 to 30 grains). 3 Double that of quinine (BUTLER, 1885). 40 to 50 grains have caused se rious effects. It appears in the urine in about two hours after its administration, and can be detected by applying the ferric chloride test to the entire urine (CARUSO, 1885). c. Dimethyloxyquinizine is very freely soluble in water, al- cohol, or chloroform ; in about 50 parts of ether. The aqueou& solution is neutral to test-papers. Antipyrine is a base of some strength, uniting with acids to form salts, from which it is set free by the alkali hydrates. d. Ferric chloride solution gives a decided red coloration, intense in solutions of 1 to 1000 parts ; the color being changed to yellow by strong acidulation with sulphuric acid. 3 Nitrous acid, as obtained by adding a little potassium nitrite and acidu- lating with dilute sulphuric acid, gives a bluish-green color in dilute, a green crystalline precipitate in concentrated, solutions characteristic of all the quinizines (KNORR). Two drops of fuming nitric acid, added to 2 c.c. of a 1 per cent, solution of antipyrine, cause a green color, and, after heating to boiling, another drop of the reagent gives a red color (Germ. rh. Commission). Tannic acid gives a white precipitate in a 1 per cent, solution. Tests for impurities The solution in two parts of water should be neutral, and colorless or faintly yellowish, free from sharp taste, and not changed by solution of hydrosulphuric acid (Germ. Ph. Commission). CINCHONICINE. See CINCHONA ALKALOIDS, pp. 91, 94. CINCHONIDINE. See CINCHONA ALKALOIDS, pp. 157-161. 4 1 The quinizines are derived from quinoline by the introduction of (NH), with additional 2H, into the quinoline molecule. The (NH) is attached to the N in the ring, this N being united to carbon by only two bonds, instead of three as in quinoline. KNORR: Ber. deut. chem. Ges. t 17, 546, 2032; Jour. Chem. Soc., 1884, Abs., 302, 1153, 1377; Am. Druggist, 13, 239, 193, 228(1884). 2 Respecting physiological and therapeutic effects, Ther. Gazette, 1885, 9, 344, 176, 517. Also FILEHNE, 1885: Zeitsch. Klin. Med., 7; Am. Druggist, 13, 193. * Pharmacopeia Commission of Germ. Apoth. Association. 170 COCA ALKALOIDS. CINCHONINE. See CINCHONA ALKALoros, pp. 161-165. CINCHOTANNIN. See TANNINS. CINCHOTINE. See p. 93. CINNAMIC ACID. See p. 69. COCA ALKALOIDS. Alkaloids of Erythroxylon Coca leaf. Cocaine, C 17 H 21 1TO 4 . The crystallizable natural alkaloid of fresh coca. Ecgonine^ C 9 H 15 1TO 3 , crystallizable. A product of cocaine by saponification, and liable, also, to be present in the leaf. Benzoyl-Ecgonine, C 16 H 19 NO 4 , crystallizable. A by-product of manufacture of cocaine from coca. (Present in the leaf ? ) Anhydride of ecgonine, C 9 H 13 ITO 2 , crystallizable. Producible from ecgonine by moderately strong sulphuric acid with heat. Hygrine, a liquid volatile alkaloid (LossEN, 1865) little known, reported to form crystallizable salts. The existence of this alkaloid is not established. Amorphous alkaloids of coca. (" Cocainoidine, Cocaicine".) Said to be obtained in preparation of cocaine. Probably present in the leaf in some conditions of this article. Not studied. Chemical constitution. Cocaine, as an easily saponifiable body, prone to split, by hydration, into ecgonine, benzoic acid, and methyl alcohol, clearly has the immediate structure of me- thyl- benzoyl ecgonine: C 9 H 13 (CH 3 ) (C 7 H 5 O)NO 3 =C 17 H 21 NO 4 . The saponification of cocaine is accomplished by an acid which takes ecgonine into combination, or by an alkali which takes both benzoic acid and ecgonine into union, or even, it is pro- bable, by digestion with water, whereby benzoyl and methyl slowly become hydroxides. But whenever the necessary condi- tions are fulfilled with any saponifying agent, the change is shown by the equation : C 9 H 13 (CH 3 ) (C 7 H 5 0)N0 3 +2H 2 =C 9 H 15 NO 3 +C 7 H 5 O . OH+CH 3 . OH. Ecgonine, by loss of CO 2 , gives the constituents of a tropine. This change, effected by distilling the barium compound of ecgonine, shows a not distant chemical relationship between COCA ALKALOIDS. 171 cocaine and the atropine group of alkaloids. And, like atro- pine, cocaine in decompositions is liable to form quite simple pyridine compounds, showing a direct relation to the pyridine series. The saponifieations of certain other well-known alkaloids, by digestion with alkali, or with acid, or with water, as stated in each instance, may be compared by the following equations. When the change is effected by acids the produced alkaloid is left in a salt ; but when by an alkali, the produced acid is left in a salt. Ecgonine unites both with acid and with alkali. C 17 H 23 NO 3 (atropine) + H 2 O=C 8 H 15 ]SrO (tropine) + C 9 H 10 O 3 (tropic acid). ^33H 43 NO 13 (aconitine)+H 2 O=C 26 H 39 NO n (aconine)-|-C 7 H 6 O 2 (benzoic acid). C 17 H 01 NO 4 (cocaine) + 2H 2 O=C 9 H 15 ;N"O 3 (ecgonine) + C 7 H 6 O 2 +CH 4 O (meth. ale.) C 22 H 23 NO 7 (narcotine) -(- H 2 O=C 12 H 15 NO 3 (hydrocotarnine) +Cio H io5 (meconine). C 32 H 49 NO 9 (cevadine) + H 2 O=C 27 H 43 NO 8 (cevin) + C 5 H 8 O 2 (methylcrotonic acid). C^H^NOii (veratrine) + H 2 O=C 28 H 45 NO 8 (verm) -f C 9 H 10 O 4 (veratric acid). C 17 H 19 NO ? (piperine) +H 2 O=C 5 H n N (piperidine) +C 12 H 10 O 4 (piperic acid). Except narcotine (and possibly piperine) the sapoiiifiable alka- loids here given are the representative medicinal constituents of the plants wherein they are found : cevadine being the most active constituent of veratrum veride, as veratrine is of cevadilla. The acids formed in the saponifications are aromatic compounds easily reduced to benzoic acid, with the exception of methylcro- tonic acid. Yield of alkaloids from coca leaf. By the process given following, Dr. Squibb obtains from well preserved lots of the dried leaves, shipped in bales, from 0.5 to 0.8 per cent, of alkaloid. Dr. Lyons obtained from the dried leaves, shipped in bales, 0.65 to 6.75, and even 0.80, per cent, of alkaloid. The alkaloidal product of these assays consists, when good leaves are taken, in the greater part of crystallizable alkaloid, though in some part of amorphous coca alkaloids. The crystallizable alka- loid is probably nearly all cocaine ; at least both ecgonine and benzoyl-ecgonine must be pretty surely left behind in each meth od of assay, by the free solubility in water and the very slight solubility in ether of both of these alkaloids. 172 COCA ALKALOIDS. It is noteworthy that all the coca alkaloids, natural or pro- duced, so far as reported, are readily soluble in water as free al- kaloids, save only cocaine itself. Also that ecgonine and ben- zoyl-ecgonine are nearly insoluble in ether, which dissolves cocaine abundantly. The solubilities are further shown here: Crystallizdble : Cocaine THE FREE ALKALOID. THE HTDBOCHLOBIDE. Water. Ether. Water. Ether. Very slight. Soluble. Soluble. Not freely. Soluble. ' Soluble. Near in sol. Near insol. Soluble. Soluble. Soluble. Soluble. Soluble. Soluble. Soluble. Insoluble. Insoluble. Ecgonine Benzoyl-ecgonine .... Amorphous : "Amorph. alkaloids." Hygrine AMORPHOUS COCAINE. Cocainoidine. Cocaicine. The qua- litative reactions and properties of the amorphous alkaloid ob- tained with cocaine in its preparation are designated by A. B. LYONS 1 as follows : The compounds are very difficult to crystal- lize. The precipitate produced in the hydrochlorate by alkalies did not crystallize at all (compare below under Cocaine, d), neither that by picric acid. In very dilute solutions (1 to 5000) gold chloride produced after some time minute prismatic crys- tals, wholly unlike in general appearance the fern-like forms from the crystallizable salt. Platinum chloride produced a few rosette- like aggregations. On evaporation the amorphous alkaloid (pro- bably not free from non-alkaloid al matter) invariably turned dark, and if the salt was evaporated quite to dryness it was found to be imperfectly soluble in water. ECGONINE. C 9 H 15 NO 3 =185 (LossEN, 1865). Crystallizes with 1H 2 O. A pyridine derivative nearly related to the tro- pines. The alkaloidal body obtained by saponification of Cocaine. It crystallizes from absolute alcohol in monoclinic prisms. Melts at 198 C., with browning, and decomposes at higher tempe- ratures. Has a slight bitter-sweet taste. It is freely soluble in water, soluble in alcohol, sparingly soluble in absolute alcohol, and insoluble in ether. In reaction it is neutral. It forms slightly crystallizable salts with hydrochloric and other acids, 1885: Am. Jour. Phar., 57, 475. ECGONINE.HYGRINE. 1 73 gummy compounds with alkalies, and a crystallizable salt with barium. The hydrochloride of ecgonine appears in a yellowish, crystalline mass, freely soluble in water and (Calmels and Gos- sin) in alcohol. Slightly soluble in alcohol (Lossen). Ecgonine platinochloride, (C 9 H 15 NO 3 . HCl) 2 PtCl 4 , is soluble in water; less soluble in alcohol. The aurochloride is soluble in water and in alcohol. Barium salt of ecgonine (CALMELS and GOSSIN, 1885) forms slender, prismatic crystals, freely soluble in water and in alcohol, slightly soluble in ether. When the barium salt of ecgonine, as obtained, with barium benzoate, by saponifying cocaine with baryta, is distilled, an isotropine (C 8 H 15 NO) is obtained (Calmels and G.) It will be observed that ecgonine, by loss of CO 2 , presents the elements of a tropine. BENZOYL-ECGONINE. C 16 H 19 NO 4 = 289. Crystallizes with 4H 2 O. Union of ecgonine with benzoic acid, the elements of H 2 O being eliminated: C 9 H 14 JSTO 3 . C 7 H 5 O. (W. MEKCK, 1885. 1 Z. H. SKRAUP, 1885. a ) Found as a by-product of cocaine man- ufacture from coca leaves. Crystallizes in transparent flat prisms. When quickly heated melts [hydrated ?] at 90 to 92 C., solidifies again and then melts [anhydrous?] at about 192 C. (Skraup). Melts, with browning, at 188.5 to 189 C. (Merck). Soluble freely in water, sparingly in alcohol, nearly insoluble in ether. It forms salts : the sulphate and acetate crystallize in. prisms. The aurochloride, C 16 II 19 ]SrO 4 . HC1 . AuCl 3 , forms yel- low scales, sparingly soluble in water, soluble in alcohol. On heating benzoyl-ecgonine with methyl iodide and an equal volume of methyl alcohol, the synthesis of cocaine is obtained : C 16 H 19 N0 4 +CH 3 I = C 17 H 21 K0 4 . HI. AN ANHYDRIDE OF ECGONINE. C 9 H 13 NO 2 . (CALMELS and GoS- SIN, 1885.) When ecgonine is heated with moderately strong sulphuric acid, an alkaloid is obtained which forms readily crys- tallizable salts both with acids and with alkalies, less soluble than corresponding ecgonine salts the barium salt having the compo- sition BaO.(C 9 Et^ 3 NO 2 ) 2 , and its hydrochloride forming stellate groups of prismatic needles. The platinochloride forms feathery groups of crystals, very soluble in water and in alcohol. HYGRINE. A volatile alkaloid found with cocaine in coca leaves (LossEN, 1865 3 ). A thick, oily liquid of a pale yellowish 1 Ber. d. chem. Oes., 18, 1594; Jour. Chem. Soc., Abs., 997. 2 Monatsch. Chem., 6, 556; Jour. Chem. Soc., Abs., 1249. Also see PAUL, 1885: Phar. Jour. Trans., [3], 16, 325. 3 W6HLER and LOSSEN: Ann. Chem. Phar., 121, 374; 133, 352. LOSSEN: "Dissertation." 174 COCAINE. color. Distils slowly with water ; distils alone between 140 C. and 230 C. It has an odor resembling trimethylamine, and a burning taste. Had no poisonous effect on rabbits. It is solu- ble in water (not in all proportions) ; freely soluble in alcohol and in ether. It unites with acids, forming salts. The hydrochloride forms deliquescent crystals. It is precipitated by iodine in po- tassium iodide solution, mercuric chloride, silver nitrate, and stannous chloride. COCAINE. C 17 H 21 IsrO 4 =303 (LossEN, 1865). Chief alka- loid of the Erythroxylon coca leaf (^IEMANN, 1860). For the yield from the leaf, and for chemical constitution and relations of the alkaloid, see above under COCA ALKALOIDS. Cocaine is identified by its effect on the tongue or eye (5), and the agreement of its precipitations (d). It is distinguished from ecgonine or beiizoyl-ecgonine by solubilities of the free al- kaloid in water and in ether (g). Its separations are effected by use of ether, etc., and/r^m coca leaf\>y several assay methods ( 11 r^ ^ ^ > S Si J3 e 3 ^ S V .2 v 5 3 S O Oi "S T^ O> ^ II C3 8 2 O """' Q *j? 2 , i ~ 3^ O 8 g 2(S i S 8 s 'S^S "Sr 2 g 2 1^. o3 02 5 00 s 3 eg 3 CO OJ " PH w OQ t 1 1 1 i 1 1 |1 f 1 03 i I "o O o 1 % H a; 1 5 1 O 1 i s 1 ff> a 03 02 'o^ .S3 j H Ls rt G 5 WATER ACID 1 S p colored solu tion. Iden residue. issolves very little. olves less thai benzene. >lves more tha benzene. ices dissolved colorless. ces dissolved. olored lilac, isiclue violet. is S'S p solves easily, sidue orange. 5 P I a P g P 1 P 03 o p 5S o CTj ^ ^ feT fe S ^ O eg tej G ^ o >. >, s > 'f. S M OQ O 1 iolves trac id on expos " ? ^ ?*g i,a r5 ^> ^3 S 'Sr 2. ^ !J It t*>fl T3 ^ rtracts on! n purities . ssolves on! traces. 1 .1 ored yello low crysta a I t>i o fi| ii 11 11 H'~ p | ID o o ' 8 i g jj - M . : OD 1 s 2 .. '>, . ,, , 1 4 j |3 'S . o m O "-H fl >j S-i EL s -2 -2'H -2 'S fe a 1 o 1 : gft s 5 1 - _ - ^3 J 1 M w 1 8^ 3 g p i ., 1 . i . d L 1 it 1 JB ^ 1 05 R i ,0 1 |1 i || 11 "oS i! s? K ^9 03 w >>"% ,J5' ^3 "S '3 crj i P J ^ 2l $ 'S r2 JU- 8 g ^ C ^jQ j OD <.: s o> g % 00 Jl s t r 1 ** 3 OD ^> Bilk 1 .3 "M 1 P 11 F 1 Colored Solution Residu Dissolves c 1 1 ll .S5 |l ^S W S P o oe ! d 1^ G E.fl OD 1 ti c e S 1 1 II Is niX! c rl me as benzi lution oranf esidue brow S t- tion yellow fluorescent. solves nothi "* ssolves trac< S H * cc * K 5 1 P P . d bi si . ^Dg S fl PC pj fe ^ c o d O *^ ^ QJ c ^.- ^'C '42 1 "o 11 || ^1 J3 "o OD 5 -^ o 11 'C-d U 1 S o> ll l| s *> ll 1 1 II S'o ? H '~~ ' ^H r^ ^ ^ % * ^-^ 'c *^ ? ' 1 1* C C ^ CO ''^ i, P P P S|" 1 I o ^ d 1 ^ ^ 1 f 1 1 ?i 1 E o 1 J Pi i i 1 i - S 1 s s i p 1 1 S 03 > 03 1 "o s 1 f | i oT ff B. Bt ll c 2 m .S ine-Orange "T3 1 1 1 I OD ne-Yellow orallin. 1 i 3 s '3 i i s .S > < <1 g : ^ '3 1 '3 'H 3 REACTIONS OF DYES. 197 s g d S g d a a ^H 2 3 2 2 g a a g ^i-S i 2 i o fc C3.2 : - l-t 2 "o eg | S '^ 5 - 3 S2 u ft ft ft a ft ft ^; OH go | -d -2 t3 s 2 K "" v> . | ^ - - -g ^ 3 3 1 ^ 2 1 IT S;E /v^ "^ 3 st i "j o * tl) ' o - * o "o * * S ft P, ft ft Tannic Acid. Not turbid. :: Brown precipitate. 3 Turbidity. Not turbid. - ;: Turbidity. Not turbid. Turbidity. 5 IN a ,=t ^ a if Dissolves deep blue. I! Dissolves yellow. Dissolves green-yellow. Dissolves yellow. Concentrated Sulphuric Add. & jj 1 No change. Dissolves brown. Dissolves brownish. Dissolves yellow. Dissolves blood-red. Dissolves dark yellow. Dissolves yellow. Dissolves, with violet precip. Dissolves yellow. e & 4* ^ a Alizarin. Aniline-Blu soluble. Aniline-Blu insoluble. Aniline-Brov (Havana- brown) . Vesuvin-Bro^ Aniline-Oran Aniline-Rec Aniline-Viol soluble. Aniline-Viol insoluble. Aniline-Yell Chrysamm Acid. Corallin. 198 ELEMENTAR Y ANAL YSIS. CONCH AIRAMINE, CONCUSCONINE. See CINCHONA ALKALOIDS, p. 92. COTTON-SEED OIL. See FATS AND OILS. CREAM OF TARTAR. See TARTARIC ACID. CUPREINE. See pp. 92 and 153. CRYPTOPINE. See OPIUM ALKALOIDS. DYES. See COLORING MATERIALS, p. 181. ECGONINE. See p. 172. ELEMENTARY ANALYSIS OF CARBON COMPOUNDS. A. qualitative analysis for the organic elements, C, H, and N, is only made for the purpose of determining whether a carbon compound be present or not, or whether a given or- ganic compound be nitrogenous or not. In the case of bodies not rapidly volatile, (1) ignition in the open air, either on platinum foil or in a glass tube open at both ends, will show carbonization in case a carbon compound be present. The fact of carbonization is shown first by the appearance of a black resi- due, and then by its gradually burning away. In the case of volatile bodies, or when for any reason the result of simply ignit- ing the body by itself proves uncertain, a resort is had to (2) igni- tion with copper oxide in a small combustion-tube, with tests of the gas evolved. The dry substance is mixed with an excess of copper oxide (previously ignited and cooled), the mixture intro- duced into a small tube of hard glass, the tube being closed at one end and fitted at the other with a tubulated cork carrying a small glass tube bent at right angles. On applying heat, very gradually, to the combustion- tube, the resulting gas is passed into lime solution or baryta solution. If a precipitate be formed this is to be gathered in sufficient abundance, and its solubility in acetic acid with effervescence is tried, for the identification of carbon dioxide. Meantime it is observed whether there be con- densation of liquid in the bent tube or not, and droplets so ob- tained may be tested, with anhydrous cupric sulphate, for water, as evidence of hydrogen. But this evidence is dependent upon the absence of moisture or hydrates in the contents of the com- bustion-tube. Unless the result of the simple test just men- tioned be clearly conclusive, it is better to use the safeguards ELEMENTARY ANALYSIS. 199 against moisture directed for Quantitative estimation of carbon and hydrogen. That is, the substance and the copper oxide are properly dried and secured from the moisture of the air, and the air in the tilled combustion- tube is replaced by dried air, before the combustion. Then the combustion is conducted very slowly, and the small conducting tube is kept cold. To be certain that carbon dioxide obtained by ignition does not come from carbon- ates that is, from non-alkali carbonates or alkali bicarbonates the material is first to be tested for carbonates. If these are present, enough of hydrochloric or sulphuric dilute acid is add- ed, and the material dried again. If it be found that a carbon compound be present, to find whether it be a nitrogenous compound or not, it is sufficient, in the greater number of cases, (3) to heat the dry substance, well mixed with dry soda-lime, when the nitrogen is given off in the form of ammonia. The heating must be to redness, and thorough drying of the material, as well as previous ignition of the soda- lime, render the operation much more convenient. An ordinary test-tube may be used for this combustion ; but a section of com- bustion-tubing, of hard glass, with one end closed, serves better. The tube may be wrapped in a strip of copper gauze near the open end, and held by the forceps, while the heat of the flame is very gradually applied. The test for ammonia is made by moist- ened red litmus-paper, also by the odor, and the color given a drop of dilute solution of copper sulphate held on a loop of pla- tinum wire. Bodies rich in nitrogen give the odor of singed hair when merely burned in the air. Heating with fixed alka- lies does not cause the production of ammonia from the nitrogen of all organic bodies. Some bodies so treated yield vaporous alkaloidal compounds, mostly showing the alkaline reaction to litmus, but not exhibiting other characteristics of ammonia. Other bodies, as many of the nitro-compounds, when treated by combustion with fixed alkali, give no indication of the presence of nitrogen. For these it is necessary, and for all it is sufficient, to (4) heat the substance with a fragment of metallic potassium for some time (SpicA, 1880), and then test the mass for cyanides. The fused mass is digested with hot water and a ferrous salt, acidulated, and a drop or two of ferric salt solution added. The blue color of ferric ferrocyanide gives evidence of nitrogen in the material taken. Also the test may be made for production of sulphocyanate by digesting the mass (after fusing with the potassium) with ammonium sulphide, and then acidu- lating. A qualitative examination for sulphur, phosphorus, sele- 200 ELEMENT A RY ANAL YSIS. nium, and arsenic may be made by applying a strong oxidizing agent, and then testing for sulphuric, phosphoric, selenic, and arsenic acids. The material (free from the acids last named) is either digested with strong nitric acid (sp. gr. 1.42) or smelted . with potassium nitrate, afterward treated with water, and the ni- trate tested for the acids. For arsenic the material may be treated, as in the examination of animal tissues for arsenic, by drying, digesting with concentrated sulphuric acid and repeated small additions of nitric acid until the carbon compounds are oxidized, and the nitric acid then wholly expelled, afterward neu- tralizing with magnesia, and subjecting the nitrate to Marsh's test for the arsenical mirror. Arsenic will sometimes be found by igniting with sodium acetate, when cacodyl* compounds are revealed by their odor. Phosphorus may usually be found by heating the carbonized material with powdered magnesium, inti- mately mixed, in the bulb of a reduction-tube, after which phos- phorescence appears in the dark. For chlorine, bromine, and iodine, as elements in an organic compound, it is necessary to effect such a decomposition as will bring the chlorine, etc., into union as chlorides, etc., or into the elementary form. Thus chloral, chloroform, and other similar compounds do not react with silver nitrate to form silver chlo- ride, etc. The necessary liberation of the haloid elements is ob- tained in some cases by digesting with strong potassium hydrate solution, in other cases by igniting in mixture with an excess of lime (each of known purity), after which the aqueous nitrate may be acidified with dilute nitric acid, and treated with silver nitrate solution for precipitates. See further upon the quantitative de- termination of the halogens. To remove organic substances, in preparation for a search for inorganic bodies in general, methods of ignition, use of oxidizing agents, application of solvents, and dialysis are described in the author's "Qualitative Chemical Analysis," third edition, para- graphs 773-778. Finally, in qualitative analysis for the elements in a portion of organic matter, instead of the direct examination for these elements, above described, the analyst will most often determine at once what organic compounds known in chemistry ^ he has in hand, recognizing their likeness by their sensible qualities, fixing their identity by well-tried qualitative reactions, resorting to ap- proved means for their separation, and proving their purity by authorized tests for this purpose. A constant boiling point and prescribed melting and congealing points are sought. The qualitative determination of a known organic compound carries ELEMENTARY ANALYSIS. 201 with it the evidence of the constituent elements of the compound. Just as qualitative tests for ortho-phosphoric acid, and for its purity, prove the presence of phosphorus and hydrogen and oxy- gen in combination as H 3 PO 4 ; so qualitative tests for benzoic acid, and for its purity, suffice to show that only carbon and hy- drogen and oxygen are present, and that these elements are united as C 6 H 5 CO 2 H. The means of separating organic compounds, and purifying them, have much in common with like means for inorganic bodies. Solvents are applied, precipitations are made, crystallization is instituted, fractional distillation is performed, chemical reactions are applied ; and these and other means, as given throughout this work, are persevered in until, in all quali- ties, constants are reached. But when in the course of research a new organic compound is obtained, and separated in purity, as shown by constant properties, it becomes necessary to find what elements it contains and in what proportion they stand. Quali- tatively, in most cases it is evident from the origin and proper- ties of the new body what elements it contains ; so that the inves- tigator may proceed at once to establish quantitatively, by the methods of organic combustion next to be described, in what proportions the elements are united, and then what molecular weight it has and under what chemical formula it is to find a place in science. Further upon the scope of qualitative and quantitative organic analysis, often termed " proximate organic analysis," and to what extent it depends upon elementary or " ultimate " organic analysis, see the article upon OKG-ANIC ANALY- SIS in this work. ELEMENTARY ORGANIC ANALYSIS, in the Quantitative Deter- mination of the Elements of an Organic Compound often termed " Ultimate Organic Analysis" rests upon the principles already outlined for the Qualitative Determination of the Organic Ele- ments. For the carbon and hydrogen a complete combustion is instituted in such a way that the combustion-products, carbon dioxide and water, are obtained as measures of these two funda- mental elements. And this simple application of the chemistry of combustion has been the means of obtaining the quantitative composition of organic bodies, from the first establishment of chemical science to the present time. 1 For nitrogen, either an 1 LAVOISIER, 1781-1784: burning of the substance with a measured volume of oxygen, and measurement of the volume of carbon dioxide produced, for cal- culation of weight : Mem. A cad. Sci., 1784-87. BERTHOLLET, 1810: Mem.de I'lnstitut National, n, 121. SAUSSURE, 1807-1814 : Ann. CMm. Phys., 62, 225; 78, 57; 89, 273. GAY-LUSSAC and THENARD, 1810-1816: use of chlorate 202 ELEMENTAR Y ANAL YSIS. ignition with fixed alkali is made to yield ammonia for determi- nation, or, more often, combustion with its products carried over heated metallic copper is made to furnish free nitrogen for measurement. The oxygen is obtained by difference. Methods for direct estimation of the oxygen have been proposed from time to time, as briefly indicated in succeeding pages, but none of them has come into actual use. The supply of oxygen for combustion is obtained as follows : (1) From copper oxide. This is either granular or in powder, coarse or fine. It is made by heating copper turnings or copper scale with nitric acid, finally to ignition, or by igniting copper nitrate prepared for the purpose. The granular form is obtained by incipient fusion. Both granulated and coarsely powdered copper oxide is to be of uniform size, by sifting, free from dusty oxide. For most uses in the combustion-tubes, the granular form moderately coarse, or that from the turnings, or the coarse powder is to be chosen, in preference to fine powder. That is, the column is to be sufficiently permeable by gases, so that it will not be necessary to have a channel over the oxide, in the tube. To intermix with the substance under analysis finely pul- verized oxide is sometimes employed, or obtained by trituration of the granular form during the intermixing. Oxide of copper, when heated, must evolve no nitrous fumes nor carbon dioxide. It is hygroscopic to a considerable extent, and in combustion for carbon and hydrogen it must be absolutely dry. For nitrogen determinations it is desirable to have it dry. It may be ignited, in a hessian crucible, short of incipient fusion, and when still warm put up in a flask with a neck a very little wider than the combustion-tube, and closed by a perforated stopper bearing a drying-tube of chloride of calcium. Also, it may, with advan- tage, be dried by ignition in the combustion-tube, in a current of dried air. This may be done when the oxide is to be after- ward removed from the tube to the flask in preparing the sub- stance for combustion, and it may with still greater advantage be done when the substance is burned in a boat. In use copper oxide is reduced to cuprous oxide or to metallic copper. With as source of oxygen and introduction of copper oxide, also the determination of nitrogen: Ann. Chim. Phys., 74, 47; Schweiger's Journal, 16, 16. DOBEREI- NER, 1816: Schweiger's Journal, 18, 379. BERZELIUS, from 1814: the use of horizontal combustion-tubes of glass. LTEBIG, 1831 : combustion with copper oxide, in detail nearly the same as " Liebig's method" sometimes employed at present: Ann. Phys. Chem. Fogg., 21, 1 (application to cinchona alkaloids). BRUNNER, 1838: oxygen gas supplied for combustion: Ann. Phys. Chem. Pogg., 44, 138. BUNSEN: intermixture with copper oxide in the combustion- tube. ELEMENT A RY ANAL YSIS. 203 the supply of oxygen gas at the close of combustion, the reduced copper is restored to oxide. Otherwise it may be restored by adding nitric acid, heating, and igniting. (2) From lead chro- mate. This must contain nothing soluble in water, and yield no carbon dioxide when heated. It fuses at a red heat. It is pre- pared by melting in a hessian crucible and pouring out upon a stone slab, when it is pulverized moderately line, sieved, and bottled for use. Or the melted chromate may be poured into water in a copper vessel, and the granulated mass collected, dried, and pulverized. It is not hygroscopic. In melting it adheres to the combustion-tube. In use it is reduced to the green chromic oxide with lead oxide. To use it a second time it is roasted, fused, and pulverized. After the second time it requires oxidation, by digesting the powder with nitric acid, drying, fusing again, and powdering. Lead chromate is em- ployed instead of copper oxide when sulphur, or selenium or tellurium, is present ; also, when very difficultly oxidizable sub- stances are in hand. Its greater efficiency as an oxidizing agent lies chiefly in its being fusible during the combustion. MAYER (1855) introduced into the powdered lead chromate one- tenth its weight of potassium dicliromate previously fused and pul- verized. This mixture serves to expel from alkalies or alkaline earths, if these be present, the carbon dioxide they may have absorbed from the products of combustion. (3) A stream of oxygen gas is employed. This is supplied most evenly and satis- factorily from a pair of gas-holders, the one filled with oxygen, and the other with atmospheric air, the stream from each being purified by passing through at least two U -tubes, one filled with pumice-stone and sulphuric acid, to dry the gas, and the other filled with fragments of potassium hydrate to remove carbon dioxide. Also, without a gas-holder, a stream of oxy- gen is obtained by generating this element, in the further end of the combustion-tube itself, from lead dioxide, heated in an air-bath to 180-200 C., or by heating mercuric oxide or po- tassium chlorate by the flame. Oxygen is sometimes generated in the combustion- tube from chlorate of potassium placed in a pla tinum boat and subjected to heat. In the preparation of oxygen for the gas-holder, chlorate of potassium, well mixed by tritura- tion with one-thousandth of its weight of ferric oxide (FRESE- NIUS), is heated over the flame in a plain glass retort not over half filled. The heat is applied very gradually, and as soon as the salt begins to fuse the retort is gently shaken. When the air is expelled the connection is made with the gas-holder. If the proportion of ferric oxide be exactly adhered to, the evolu- 204 ELEMENT A RY ANAL YSIS. tion of gas will not be impetuous. 100 grams of the chlorate will yield about 27 liters of oxygen. Oxygen gas is tested for chlorine by passing it through silver nitrate solution, and for carbon dioxide by passing through lime solution. A splinter of wood which has been kindled and blown out should burst into a flame when introduced into a stream of oxygen gas. The soda-lime used as the fixed alkali, for the conversion of organic nitrogen into ammonia in the combustion-tube, 1 is a mixture of two parts of calcium hydrate with one part of sodium hydrate. It is usually made by the evaporation of a solution of sodium hydrate with the proportional quantity of slaked lime. S. W. JOHNSON (1872 2 ) recommends, as more convenient and even better, a mixture of equal parts of crystallized sodium car- bonate and slaked lime, prepared by evaporating the mixture. 3 Soda- lime is obtained in granular form, more convenient for the greater part of its uses than the powdered form. It should not evolve any trace of ammonia when heated with sugar ; it should not be more than slightly moist ; and (unless prepared upon Johnson's direction) should not effervesce very much upon the addition of acids. It is made ready for use by igniting in a hessian crucible at a gentle heat, and while warm it is put up in a well-corked bottle, or a bottle with a tubulated stopper carrying a drying tube containing both calcium chloride and a little gran- ulated soda-lime. Metallic copper is used, while heated, to reduce oxides of nitrogen in the combustion-tube, this being necessary, first, to prevent error in estimating carbon by the absorption of carbon dioxide ; second, to avoid loss of nitrogen in estimating this ele- ment by its volume when free. Coils of copper gauze or foil, or spirals of copper wire, are heated to redness in the air long enough to oxidize the surface, and then heated in a stream of hydrogen to reduce the oxide formed. For the reduction the coils are introduced into a combustion-tube having a tubulated stopper at each end, and a current of hydrogen passed through 1 VARRENTRAPP and WILL, 1841: Ann. Chem. Phar., 39, 257. 2 Am. Chemist, 3, 161; 1879: Am. Chem. Jour., i, 77. 3 '* Equal weights of sal-soda, in clean (washed) large crystals, and of good white and promptly-slaking quicklime, are separately so far pulverized as to pass holes of T V inch, then well mixed together, placed in an iron pot, which should not be more than half filled, and gently heated, at first without stirring. The lime soon begins to combine with the crystal water of the sodium carbonate, the whole mass heats strongly, swells up, and in a short time yields a fine pow- der, which may be stirred to effect intimate mixture and to dry off the excess of water, so far that the mass is not perceptibly moist, and yet short of the point at which it rises in dust on handling. When cold it is secured in well closed bottles or fruit-jars, and is ready for use" (where last above cited). ELEMENTARY ANALYSIS. 205 until the air is expelled, when heat is applied as the stream of hydrogen continues. Coarsely granulated copper oxide, reduced by ignition in a current of hydro- gen, is employed to some extent instead of the spiral coils, and is more efficient than they. All copper reduced by ignition in a stream of hydrogen is liable to contain traces of occluded hydro- gen, from which error may arise unless precaution be taken. 1 At ordinary temperature it quickly absorbs moisture from the air. Copper gauze and wire are also used in the combustion-tube in methods of combustion of non-nitrogenous bodies, requir- ing only to be cleaned by a mo- mentary ignition in the clear flame before use. Solution of Potassium Hy- drate. To absorb carbon dioxide in potash bulbs, good potassium hydrate nearly free from carbo- nate is dissolved in an equal weight of water. Some chemists use a solution in 2 parts of water ; others a solution in f part of water. The solution dropped into diluted mineral acid should not effervesce. It should be strictly free from nitrite. It is sometimes used a second time. Solid hydrate of potassium is also employed for absorption in elementary organic analysis, taken either in stick or in lump, the drier the better. Chloride of Calcium. For absorption of the water resulting from combustion, dried calcium chloride strictly free from alka- 1 G. S. JOHNSON, 1876: Jour. Chem. Soc., 29, 178. 206 ELEMENTARY ANALYSIS. line reaction is employed. In preparation the solution is stirred while evaporating, to granulate, and the residue dried at about 200 C. It consists of CaCl 2 .2H 2 O. The granulated form is much preferable. It may be tested, in concentrated solution, with litmus-papers. It may be prepared from crude fused cal- cium chloride by dissolving in lime solution, filtering, neutraliz- ing with hydrochloric acid, evaporating to dryness, and heating as above directed. But to be well assured that the calcium chlo- ride is free from uncombined bases, the operator should take the precaution to pass dried carbon dioxide through the filled chlo- ride of calcium tube for an hour or two, and then a current of dried air to restore the normal weight of the tube. For drying gases the crude, fused calcium -chloride, in broken masses, is all that is required. It usually has an alkaline reaction. Combustion- tubing is to be of hard potash-glass, mostly of 12 to 14 millimeters ( T \- to J inch) inner diameter, and about 2 millimeters (not quite j- inch) thickness of glass. It is best obtained in lengths sufficient for two tubes that is, in pieces mostly 5-J- to 6J- feet long. For many purposes the combustion- tube is drawn out at one end, and preferably in bayonet form, as in Fig. 9. A section of tubing long enough for two combustion- 6 H.g. 9 c cf tubes is readily so drawn and bent that when severed in the cen- tre the two finished tubes are obtained. The ed^es are to be rounded in the flame. A combustion-tube is cleaned with a piece of muslin or paper attached to a stiff wire, and is dried by heating over a flame or on a water-oven, while from time to time the air is drawn out through a small tube carried in to the closed end, when it is well stoppered. Combustion-tubing of glass not sufficiently infusible may be used by wrapping it with copper gauze. Iron tubes are some- times used, with special precautions, especially for nitrogen de- terminations by ignition with the soda-lime (CLOEZ, 1863 ; JOHN- SON, 1879). A hard-glass tube may be used repeatedly for combustion in a stream of oxygen gas, and sometimes more than once for combustion with admixture of the substance with oxide of copper, not more than once for combustion with chromate of lead. Chloride of Calcium Tubes, for the absorption of the water of combustion and for drying gases, are used of various patterns. ELEMENT A RY ANAL YSIS. 207 including the one-bulb and two-bulb straight tube, and the U-tube with and without a bulb : Fig. 10, and in position in Figs. 8 and 16. The tubulated stoppers should be of rubber, or cork waxed over. An empty bulb in the horizontal part of the chloride of calcium tube has the advantage that it serves as a cup for a por- tion of the water which condenses in it, and the chloride of calcium the longer retains its power of absorption. A tuft of cotton-wool is drawn into the tube, so as to rest firmly against and within the narrow part of the tube through which the current enters, when the fragments of calcium chloride are filled in, and at the other end a cover of cotton- wool or muslin is placed. Concentrated Sul- phuric Acid has been variously used, instead of calcium chloride, to absorb the water. 1 Potash bulbs are of the two principal patterns, GEISLER'S, Fig. 11, which are to be preferred, and LIEBIG'S, Fig. 12, which have long been used. When in use the larger bulb is placed next the combustion-tube. In being filled, the end which is nearest the combustion the one into which the stream of gas is . 11 to enter is inserted into the solution of potassa, and a sufficient amount of the liquid is drawn into the apparatus. The proper quantities of potash solution are shown in the figures. Instead of a bulb apparatus for potash solution a large bulbed U-tube, illled with soda-lime, is sometimes used as an absorbent of the carbon dioxide of combustion. A potash tube, either straight or 'DIBBITS, 1876: Zeitsch. anal. Chem., 15, 122; MORLEY, 1885: Am. Jour. Sci., [3]. 30, 140; Chem. News, 54, 33. 208 ELEMENTARY ANALYSIS. U-form, filled with fragments of dry potassium hydrate or with granulated soda-lime, is used beyond the potash bulbs, and weighed with it. It guards against loss of water- vapor and of traces of carbon dioxide. The Combustion -Furnace of ERLENMEYER is shown in Fig. 8, p. 205. It requires a good supply of gas. The combustion- furnace of G-LASER, preferable for some combustions, is shown in Fig. 16. In the use of a gas combustion-furnace the supply of air must be regulated with that of gas to each burner. The furnace should be placed where it will be secure against currents of air or the access of acidulous or arnmoniacal gases. THE CONDITIONS OF SUCCESS in organic elementary analysis are attained by a watchful attention to details, with a faithful study of the sources of error, throughout the operation and in the preparation for it. The sources of error are so many that even an experienced operator, when commencing work with newly collect- ed appliances, is quite liable to failure. When the work is well in hand, and operations upon material of known composition are made to succeed each other with almost invariable success, an important estimation may be undertaken with confidence in the result, but this is to be obtained as the mean of several nearly coinciding determinations. ESTIMATION OF CARBON AND HYDROGEN IN BODIES NOT CON- TAINING NITROGEN. Oxygen supplied by Copper Oxide. Analy- sis of Solids. The substance to be analyzed, obtained of exactly constant composition, in respect to hydration and freedom from all foreign matters, and (if pulverizable) in very fine powder, is introduced into a small weighing-tube a light cylindrical con- tainer, with a caoutchouc or fine cork stopper, and of 3 to 6 c.c. capacity. For each elementary estimation from 0.3 to 0.4 gram is usually taken, and estimations may require repetition ; therefore it is better to take from 2 to 4 grams of the sample at once in the weighing- tube, so that all the desired estimations can be made upon material of constant composition, without danger of loss or gain of moisture or other constituents. When it is desired closely to regulate the quantity of substance for each combus- tion, it is well to employ in addition a smaller weighing-tube to receive enough for one combustion, which is transferred from the larger weighing-tube. The management of liquids, soft solids, and very volatile matters is given hereafter (p. 213). The charging of the combustion- tube, under whatever order of arrangements, is to be so effected that the entire contents of the tube including the substance under analysis, the material ESTIMA TION OF CARBON AND HYDROGEN. 209 supplying oxygen, oxygen gas, and atmospheric air shall be strictly free from moisture before the combustion begins. To remove moisture and exclude it from the materials and the air entering into the combustion-tube, different orders of operation are adopted in different laboratories and directed by different authorities. When the substance is not burned in a boat of platinum or porcelain, and when the oxygen is supplied by copper oxide, the work may be conducted as follows : The filled potash bulbs, dried with filter-paper at the ends arid wiped clean, with the attached potash tube (if this be employed), are weighed, and both openings are afterward closed with sections of clean rubber tubing stopped with a bit of glass rod. The chloride of calcium tube is weighed, and its ends afterward closed. The weighing-tube, narrow and of considerable length, containing the substance for analysis, is weighed without opening it. There is provided granulated oxide of copper, which has been taken after ignition, and while warm, into a filling-flask. 1 as described on p. 202. The dry combustion- tube, with its drawn-out end sealed, is rinsed with some oxide of copper. About four inches (10 centimeters) of the body of the combustion-tube is filled with the oxide of copper, taken from the flask by the mouth of the tube. The substance is added, upon the layer of copper oxide, from the weighing-tube, which is in- troduced into the combustion-tube, avoiding the adhering of the substance to the inner surface. The weighing-tube is closed and put aside to weigh again. Another layer of oxide of cop- per equal to the first is taken into the combustion-tube, add- ing at first in such a way as to rinse the latter. With a stiff iron wire as long as the combustion-tube, bent in a single cork- screw turn at one end and in a ring at the other (Fig. 8), the substance is well mixed with the oxide of copper, leaving undis- turbed about 4: centimeters (1-J- inches) of the layer of oxide next to the bent end. Oxide of copper is added to fill to within about 6 centimeters (2J- inches) of the mouth. A porous plug of as- bestos is added, leaving a good free space, to be kept clear of condensed water, between the asbestos and the tubulated caout- chouc stopper. If a cork stopper be used less space is required. Another method of charging the tube, when copper oxide is the sole source of oxygen for combustion, provides for mixing 1 The copper oxide may be dried by ignition in the tube with advantage in this method as in others. The tube is filled with the oxide, then the open drawn-out end is connected with a set of drying-tubes, and dried air is either sent by a gasometer or drawn by an aspirator through the drying-tubes and the oxide of copper, while the latter is ignited. 210 ELEMENTARY ANALYSIS. the substance with some of the oxide of copper in a mortar of glass or unglazed porcelain. The warmed mortar is placed on a sheet of glazed paper on the table, and the oxide of copper is taken warm. Both the tube and the mortar are rinsed with some of the oxide of copper, and the rinsings put aside to be ignited again. After a layer of about an inch (2 centimeters) of the oxide of copper next to the bayonet-end of the tube, a mix- ture of the substance with oxide of copper is made by gentle tri- turation in the mortar, and added in such quantity with the mortar rinsings as will fill the tube to or a little beyond the mid- dle of its body. The remainder of the tube is filled with the copper oxide to within about 2J inches (or 6 centimeters) of the mouth, covering with a porous plug of recently ignited asbestos. When the contents of the tube are in fine powder a channel for the easy passage of gases is made by tapping the tube upon the table as it lies in horizontal position. With granulated cop- per oxide, or that in coarse powder, a channel is usually to be avoided. The removal of atmospheric moisture from the filled com- bustion-tube, when a gaseous supply of oxygen is not used, may be accomplished by attaching a drying-tube of chloride of calcium, and repeatedly pumping out the air, which is each time permit- ted to flow back through the drying-tube. A small exhausting- syringe may be used, or a filter-pump acting through a flask provided for the admission of air at will. But it is a more satis- factory way to pass a current of dried air, drawn by an aspirator or sent by a gasometer, through the tube from the drawn-out end, as directed farther on to be done for another purpose after the combustion (p. 212). When the contents are dried the com- bustion-tube is kept closed by a caoutchouc stopper until con- nected with the weighed chloride of calcium tube and potash apparatus for the combustion. Chr ornate of lead (p. 203) is used instead of oxide of copper for substances difficultly oxidizable, as well as when sulphur is present. In the charging of the tube it is used in the same man- ner as oxide of copper. Having a higher oxidizing power than copper oxide, a smaller quantity is required, and a narrower tube may be used. The contents of the tube should be dried the same as when oxide of copper is used. Bichromate of Potassium, with Oxide of Copper, may be used as follows (GINTL, 1868): The combustion- tube is charged, first, with about 2J- inches (6 centimeters) length of granulated copper oxide ; then with about 1J inches (3 centimeters) length of acid cliromate of potassium which has been fused, pulverized, and ESTIMA TION OF CARBON AND HYDROGEN. 211 kept dry ; then the substance added from the weighing-tube, and again oxide of copper to make about 1J inches (3 centimeters). With the mixing wire (p. 205) the substance is well mixed, leaving undisturbed about half of the layer of copper oxide next the bayonet-end. The tube is filled with copper oxide ; an as- bestos support is placed, providing an open space next the tubu- lated stopper ; and the contents of the, tube are deprived of moisture, as before directed. In making the combustion with oxide of copper, the com- bustion-tube is placed in the furnace with the end next the chlo- ride of calcium tube projecting as far as the asbestos plug. A disc of copper foil may be employed as a shield over the tube to protect the stoppered end from too great heat. The tightness of the apparatus can be assured by expelling a little air, by heat- ing the bulb of the potash apparatus nearest the combustion-tube until a few bubbles of air have escaped, when the liquid rises on the side heated and should then remain stationary. If the rubber connecting tubes are not snug they are bound with wire. The oxide of copper next the chloride of calcium tube is heated first, very gradually, to dull redness, and the heat is steadily carried toward the substance, not rapidly enough to cause a tumultuous escape of expanded air through the potash bulbs. At the end near the mouth the combustion-tube is maintained uniformly at a temperature high enough to prevent the condensation of water- vapor within, but not high enough to endanger melting the tubulated stopper if of caoutchouc, or charring it if of cork. The column of copper oxide, back to where the combustible substance begins to be intermixed, is held at dull red heat, not high enough to endanger blowing-out of the glass, while now the heat is carried very gradually back through the substance itself so gradually that not more than one or two bubbles a second will pass the liquid in the potash bulbs. Certainly the bubbling should not at any time be too rapid to be counted. There should not be ernpyreumatic odor in the escaping air. When the air has been nearly all expelled, and the gas which passes out of the chloride of calcium tube consists mainly of carbon dioxide, the bubbles will pass through the last potash bulb only at conside- rable intervals, and these intervals will be longer if that portion of unmixed oxide of copper back of the substance be heated, as it may be \n part and with caution, before the substance begins to burn. At the end of the operation all the contents of the tube are held at full heat. As the current of carbon dioxide ceases the liquid in the potash bulb next to the combustion-tube rises. Slight suction may now be applied to the potash tube. 212 ELEMENTARY ANALYSIS. At this time, or in anticipation of the time when the combustion with the copper oxide is completed, the heat is turned off under the rear end of the combustion-tube so that the drawn-out extre- mity is cooled, and this is then connected by a rubber tube with a set of tubes for thoroughly depriving air or oxygen of moisture and carbon dioxide. Such* a set of tubes is described, together with the means of supplying oxygen and air, in the directions for combustion in a current of oxygen gas, following. The pot- ash tube is connected with an aspirator, either the bell jar form shown in Fig. 16, or a bottle aspirator, serving not only as a pump but as means of regulating the flow of gases supplied, and of preventing recession of the current. To remove now the carbon dioxide and water-vapor in the combustion-tube, and at the same time insure the absolute com- pletion of the combustion, if oxygen gas has been provided, it is better to pass purified oxygen gas from the connection at the bayonet-end (Fig. 13) through the combustion-tube while it is g n E W L ___i UUU UUUUliUUUUUUUU Fig. 13 heated. The connection is opened by breaking the point of the combustion -tube, in the rubber-tube, with a pair of pliers, and a sufficient stream of oxygen is passed. Now, as oxygen has a higher specific gravity than air, the former is to be removed from the absorption-tubes to be weighed, by washing it out with a stream of purified air. This is done by changing the connection from the oxygen gasometer to an air gasometer (the position of which is shown in Fig. 13), taking air through the same tubes for depriving it of carbon dioxide and moisture. Or, without a gasometer for air, the previous connection with the oxygen sup- ply may be opened for the admission of air, purified as just stated, and drawn through by the aspirator, long enough to re- move the oxygen. As soon as the stream of air is applied the heat may be diminished, turning it down very gradually to avoid the breaking of the combustion-tube. Without oxygen gas, air ESTIMATION OF CARBON AND HYDROGEN. 213 dried and purified as above directed may be drawn through the combustion-tube while it is maintained at full heat, until the carbon dioxide is removed from the apparatus. The combustion of a substance mixed with copper oxide and with a stream of oxygen throughout the operation, as sometimes done, can be readily understood from the directions foregoing, together with those given in the following pages upon Combustion in a Pla- tinum Boat with gaseous oxygen. Again, some operators, with the benefit of experience, merely break the point of the com- bustion-tube in the open air, and draw through, by the aspirator or by the mouth, sufficient air to displace the gaseous content of -the apparatus, as indicated by the bubbles no longer diminishing in size as they pass the potash bulbs. The chloride of calcium tubes, and the potash bulbs with the potash tube, are at once closed with the caoutchouc caps, and are weighed without these additions. With lead chromate the combustion is conducted so that the chromate between the substance and the mouth of the tube is not fused, but remains porous. The lead chromate intermixed with the substance is not fused at first, nor until the substance has all been heated ; but it should be wholly fused at last, because it is a much more powerful oxidizing agent in the liquefied state. The errors to be guarded against in combustion with oxide of copper or chromate of lead are those of too high figures for hydrogen and too low figures for carbon. With dry potash in the end tube, the use of an aspirator, and a stream of dry air to recover the carbon dioxide left in the apparatus at the close of the combustion, the loss of carbon may be avoided. To prevent an excess "of hydrogen requires vigilance, its accomplishment lying mainly in the absolute removal of moisture before combus- tion. It has been stated 1 that without the potash tube the carbon averages about 0.1# too low, while with the potash tube it averages near 0.05^ too high ; and that [without the substitution of dried air in the filled combustion-tube] the hydrogen averages 0.1 to 0.2$ too high. Liquids are weighed and introduced into the combustion-tube in glass bulbs. For volatile liquids these may be made by draw- ing out wide tubing, Fig. 14, the drawn-out portion being about 5 millimeters ( T \- inch) in external diameter, and in the wider portion about 3 centimeters (1 inches) long. For either volatile or non-volatile liquids bulbs of the shape shown in Fig. 15 1 KBKULfi's " Organische Chemie." 1867, i. 22. 214 ELEMENT A R Y ANAL YSIS. may be employed. Bulbs are filled by passing through the flame to heat the air they contain, and then immersing the open end in the liquid, which presently rises to fill part of the tube. If the liquid be volatile, it may now be made to boil in the tube, when, the open end being inserted in the liquid, an additional quantity is obtained. If an open bulb be placed with its mouth uirler the surface of a liquid, and the whole put under an air-pump, on drawing out the air the liquid rises afterward in its place. Non- vola- tile and slightly volatile liquids are weighed and introduced into the com- bustion-tube in 'ftpen bulbs ; freely p. ,, j volatile liquids are weighed in sealed ^3^ /J^Fig.15 bulbs. In any case the weight of the empty bulb is taken before filling ; and the capillary neck of the bulb is drained as fully as possible after filling. To seal the mouth it is held a moment in the flame, and when cool it is ready to be weighed. The combustion of non- volatile liquids and soft solids is much better done with a stream of oxygen gas, in a platinum boat. The products of destructive distillation are burned almost as fast as formed, the substance itself being heated very gradually. On the other hand, when freely volatile bodies are burned in oxygen gas, care is required, owing to some liabi- lity of explosion in the combustion- tube. The use of oxygen gas to complete the combustion of the carbonaceous residues of volatile substances is, however, always desirable. And WARREN* has presented a method of burning volatile bodies with oxygen gas, by means of a combustion-tube packed with asbestos, the heat being applied and the combustion effected only in the an- terior end of the tube, while the substance is vaporized in the posterior end. A long combustion- tube is used, and the column of porous asbestos packing acts like the gauze of Davy's safety- lamp. In filling the combustion- tube, when liquid or volatile bodies are to be burned with copper oxide, the coarsely granular oxide is taken, a layer of about two inches of the same is placed at the posterior end, the substance contained in two bulbs is in- troduced with some copper oxide between them, while the com- bustion-tube is upright, and the tube is filled up with copper oxide. If the bulbs have been sealed, a file-mark is made upon the neck, which is broken as the bulbs are dropped into the tube. Yery volatile substances are sometimes introduced in small por- '1864: Chem. News. Zeitsch. anal. Chem., 3, 272. ESTIMATION OF CARBON AND HYDROGEN. 215 tions, in several very thin bulbs, which, by holding a hot clay shield near, are made to burst in the filled combustion -tube, either while only copper oxide in the front is heated, or before heating 216 ELEMENTAR Y ANAL YSIS. at all. Less volatile liquids, introduced in open tubes, may be intermixed with the oxide of copper by applying a single stroke of the exhausting syringe to the tilled combustion- tube, causing the liquids to boil. Combustion-tubes of good length and width are required, with evenly coarse granular copper oxide filling the tube without a channel. Care is exercised to avoid explo- sions and the escape of unburned vapor. It is desirable to shield the combustion-tube under a firm cover of copper gauze. Gaseous bodies are subjected to the special methods of Gas Analysis for elementary estimations. These methods depend most- ly upon volume measures of the gases, with measures of the residues after their absorption, and the products of their combustion. Such a volume measure of the residue after absorption is made in the chief method of the analysis of solids for nitrogen, as described in the pages following. In the first elementary analysis of fixed bodies, by Lavoisier, the products of the combustion were mea- sured in volume for the calculation of weight. Methods of or- ganic analysis for carbon, founded on gas measurements, have been reported upon by SOHULZ (1866) and others. Gases may be subjected to the method employed for the relative determination of the carbon and nitrogen of fixed substances, by volume mea- surement after combustion, as devised by Liebig, Bunsen, Mar- chand, and others. The combustion in a platinum boat, with gaseous oxygen and copper oxide, may be conducted, for a non-volatile substance, as follows : The furnace should have a secure, level, concave support for the combustion-tube. The furnace of GLASER (Fig. 16) has gutter-shaped iron supports, which may be placed together to form a continuous canal. The combustion-tube, of 12 or 14 mil- limeters (near \ inch) internal diameter, and preferably 4 or 5 centimeters (1-J or 2 inches) longer than the furnace, is open at both ends, with fused edges and" tubulated rubber stoppers. The platinum boat is of size to easily enter the tube. The oxide of copper, granulated, is taken cold. Copper gauze and wire are provided, the gauze in pieces about 2 centimeters (} inch) wide, rolled in plugs large enough to fit with easy friction in the com- bustion-tube, and cleaned by momentary ignition in a Bunsen flame. One of these plugs is pushed about 25 centimeters (10 inches) into the tube ; from the other end the coarsely granular copper oxide is filled to within 6 to 8 centimeters (2 to 3 inches) of the opening, settling it by very slight tapping, following which is inserted another plug of the copper gauze of sufficient length, leaving a free space between it and the rubber stopper. 1 1 A spiral of copper wire is used, forward of the plug, by some chem- ESTIMATION OF CARBON AND HYDROGEN. 217 A shield of copper foil is put over this end (Fig. 16). A piece of copper gauze about 10 centimeters (or 4 inches) wide is rolled about a stiff copper wire of sufficient length, doubling a bit of the wire down firmly upon the first turn of gauze, and rolling the gauze to make a plug to fit the tube easily, when the free end of the wire is bent, forming a ring which will enter the tube, in which it is placed after igniting it for a moment. The gasome- ters for oxygen and for air are filled, and connected with an apparatus for removing moisture and carbon dioxide. Each gasometer may be connected wifh a separate bottle of potassium hydrate solution, from which both connections may lead to a single deep U-tube filled with coarsely granular soda-lime, and then successively to three deep U-tubes filled with small lumps of dry fused calcium chloride. A U-tube containing pumice-stones wet with concentrated sulphuric acid may also be interposed at any point after the soda- lime. See Fig. 16. A mercury- valve (Fig. IT) ^> Fig.17 is sometimes interposed between the coinbustion- tube and the purifying apparatus to prevent dif- fusion of products of combustion backward. A good chloride of calcium U-tube, with bulb on the horizontal part next the combustion, is filled ; also the Geisler potash bulbs (Fig. 11) with the potash tube ; and a bell-jar as- pirator (Fig. 16) is provided, carrying a chloride of calcium tube. The apparatus being put in place with the combustion-tube over the furnace, without the platinum boat, the tube is heated up throughout, and a slow current of the dry air is transmitted through the combustion-tube alone. Meanwhile the calcium chloride tube and the potash bulbs and tube are weighed with- out their caps, and then closed. When the column of copper oxide has been heated for ten or fifteen minutes the heat is turned down, the platinum boat is ignited and then cooled in a desiccator and weighed, and from 0.3 to 0.5 gram of the substance is trans- ists. All the metallic copper becomes coated with copper oxide during the heat- ing in the stream of oxygen or air, and the copper oxide so formed makes an efficient oxi'dizing agent for the* gaseous products of incomplete combustion. However this anterior end of the tube be filled, it is advisory to have a free space of 2 or 3 centimeters (an inch or more) next the caoutchouc stopper. 218 ELEMENTAR Y ANAL YSIS. ferred to the boat. The weight may be taken in the boat, or, if the substance be affected in any way by exposure, the substance is added from a stoppered tube, weighed before and after it is taken (p. 208). The air-current is stopped ; the chloride of cal- cium tube and the potash bulbs and tube are securely connected by caoutchouc tubes of clean inner surface, and the aspirator is connected in place. The stopper at the posterior end of the combustion-tube is taken out and the copper-gauze cylinder with- drawn, the platinum boat is inserted in its place near the short copper-gauze plug, the cylinder and posterior stopper replaced, and the connections made with the purifying apparatus and gasometers. The aspirator- valve is opened a little, a few burners nearest the chloride of calcium tube lighted and gradually turned up, and the heat increased to dull redness, not sufficient to distort the tube, and extended back to a safe distance from the gauze plug governing the aspirator to take out the expanded air. The diminished gaseous tension within the apparatus tightens the connections. A difference of 12 to 15 centimeters (about 5 inches) in water level of the bell-jar aspirator is usually main- tained. The gauze cylinder is now gently heated, and at about this point the stream of air may be exchanged for one of oxygen, running at first not faster than a bubble every two seconds. The space next the anterior stopper is kept dry without softening the rubber, and the heat is brought back to within 4 or 5 centimeters (1J or 2 inches) of the platinum boat, when a gentle heat is turned up directly underneath the substance. The progress of the combustion is observed, and the heat so regulated by the changes in the substance and the bubbling in the potash bulbs as to obtain a gradual and even progress. When the substance is completely charred, and the bubbling through the potash solution abates, the heat under the boat is increased and the How of oxygen quickened to about one bubble per second. The exchange of oxy- gen for air may be delayed till the substance is charred. When the carbonaceous matter in the boat has disappeared, the heat underneath it is lessened and the stream of oxygen quickened ; soon after which the heat is partly turned down all along the tube, and the stream of oxygen exchanged for one of air. In a few minutes now the gasometer and aspirator may be shut off, and the potash bulbs and tube and the chloride of calcium tube at once detached, closed at their openings, wiped, and weighed (without their caps). The platinum boat may be weighed for estimation of ash. The combustion-tube is cooled very gradually, and is at once ready for another combustion, with the same copper oxide, free from moisture. The water in the bulb of the chloride ESTIMA TION OF CARBON AND HYDROGEN. 219 of calcium tube is examined as to its purity, freedom from empy- reuma, etc. Liquid substances are weighed in bulbs or small tubes, as described on p. 213, placed upon the platinum boat, and subjected to combustion as above directed. Volatile substances are expelled from the bulbs containing them before the posterior portion of the copper oxide is heated, a hot clay shield being held over the boat for that purpose. The relations of these substances to elementary analysis have been stated further on p. 214. ESTIMATION OF CARBON AND HYDROGEN IN NITROGENOUS COM- POUNDS. The presence of nitrogen requires only such a change in the conditions of the combustion as shall prevent acidulous oxides of nitrogen being formed and carried into the potash bulbs to increase their weight. This is done by passing the products of combustion over metallic copper at red heat. The preparation of copper for this purpose is described on p. 204. In combustion of nitrogenous compounds with copper oxide, as directed on pp. 208, 211, the combustion-tube is to be 12 to 15 centimeters (about 5 inches) longer than required for a non- nitrogenous body. A roll of copper foil about 12 centimeters (near 5 inches) long is prepared as directed on p. 217, heated in hydrogen gas (p. 204), and placed in a drying-oven at 100 C. The combustion-tube is filled in the ordinary way, leaving room for the gauze roll, which is introduced while warm from the drying-oven. Before the mixture of substance and copper oxide is heated in the tube the metallic copper is brought to a bright red heat, and so maintained during the combustion. If gaseous oxygen be supplied at the close of the operation, it is supplied sparingly, so as not to oxidize all the metallic copper until, near the close of the combustion, the nitrogen shall have been expelled, and only carbon remain to be burned. In combustion of nitrogenous bodies, for carbon and hydro- gen, in a stream of oxygen gas, the gauze copper roll of about 12 cm. length, as above described, is inserted in a space left for it in the anterior end of the combustion-tube (p. 216), chosen longer on this account. The copper oxide is first dried, in the heated tube, in a stream of dry air (p. 202) ; then the air is turned off, the roll of metallic copper warm from the drying-oven intro- duced into its place, the platinum boat with the substance inserted, the connections made, and the combustion commenced. The stream of air is not changed for one of oxygen until the continu- ance of the combustion demands it ; and neither is used in such excess that the metallic copper becomes oxidized before the nitro- 220 ELEMENTAR Y ANAL YSIS. gen has all passed out. To burn out the last traces of carbona- ceous residue the stream of oxygen may be used freely. The roll of metallic copper is used but once. ESTIMATION OF NITKOGKEN IN CARBON COMPOUNDS. Absolute determination by volume of the gas. Of various serviceable methods for this estimation, the following are here presented : Method of JOHNSON and JENKINS/ based in good part upon Dumas's Method. The substance is burned in mixture with copper oxide, and, by help of oxygen generated from potassium chlorate, put in the rear of the combustion-tube, the gaseous pro- ducts being all carried through a porous column of heated metal- lic copper of length sufficient not only to deoxidize nitrogen oxides but to absorb all the excess of oxygen. A short layer of heated copper oxide, front of the metallic copper, oxidizes any hydrogen held occluded by the metallic copper, also traces of carbon monoxide formed by the metal. The gases are received in a measuring-tube (azotometer), over potash solution, which they pass through, and which absorbs all carbon dioxide, nitrogen being left alone as a permanent gas, measured for quantity. Between the combustion-tube and the azotometer is introduced a mercurial air-pump, by which the combustion-tube is first fully exhausted of air before the combustion, and by which the gaseous products left in the tube after combustion are drawn out and delivered to the azotometer. 8 During the combustion the gasfc& pass through the pump to the azotometer. After the initial ex- haustion of the combustion-tube, carbon dioxide is generated in it by heating a short column of sodium bicarbonate placed in the very front of tlie tube, this carbon dioxide, like that formed in 1 S. W. JOHNSON and E. H. JENKINS, 4880 : Am. Chem. Jour., 2, 27; Zeitsch. anal. Chem., 21, 274; Chem. News, 47, 146. A valuable report on Prof. John- son's method is given from continued experience in its use, in comparison with the Ruffle Method, by C. S. DABNEY, JR., and B. VON HERFF, 1885: Am. Chem. Jour., 6, 234. Also, valuable improvements in the pump, and a modification of the charging of the combustion-tube by T. S. GLADDING, 1882 : Am. Chem. Jour., 4, 42 (illustrated). The "' Official Methods of the Association of Agri- cultural Chemists for 1886-7" are given in Bulletin No. 12, Department of Ag- riculture, Washington, 1886, p. 52. Modifications of Dumas's Method are also given by G. S. JOHNSON, 1884: Chem. News, 50, 191; Jour. Chem. Soc., 48, 189 ; and by ILINSKI (with ordinary laboratory apparatus), 1884 : Ber. d. chem. Ges., 17, 1347; Zeitsch. anal. Chem., 24, 76. 2 DABNEY (see last foot-note) says : ' ' For getting the air, before combustion, and the nitrogen afterward, out of the tube, we have used carbon dioxide with- out a pump and have obtained excellent results. . . . Magnesite or manganese carbonate, put in the back end of the tube, are the best sources for this pur- pose. [See, following, SIMPSON'S Method.] But more time is consumed in this way than with a good, fast- working, tight pump." ESTIMA TION OF NITROGEN. 221 combustion, being taken up in the azotometer by the potash solution. The copper oxide is directed to be made by heating copper scale with 10 per cent, of potassium chlorate and enough water to make a thin paste, stirring till dry. and igniting until the mass does not glow when stirred. The potassium chloride is to be washed out by decantation, and the copper oxide dried and mode- rately ignited. Metallic copper is used as fine copper gauze in rolls to fit the combustion-tube, or as granular oxide of copper reduced and cooled in a stream of hydrogen (p. 205). JPotassium chlorate is prepared by fus- ing the commercial article in a porce- lain dish and pulverizing when cold. Sodium bicarbonate is used, and must be free from organic matter. Solution of potass a is made by dissolving com- mercial potash in sticks in less than its weight of water, and permitting the excess to crystallize out when cold. The same solution may be used a num- ber of times. The combustion-tube, of best hard glass, should be about 28 inches (71 centimeters) long. The rear end is bent and sealed as in Fig. 20. It is best to protect the horizontal part with thin sheet copper or copper gauze, as directed further on. The azotometer. Fig. 18, is a modi- fication of SCHIFF'S/ The gas is mea- sured in an accurately, calibrated bu- rette, A, of 120 c.c. capacity, graduated to fifths c.c., and closed at the upper end by a glass stop-cock. The lower end is connected, by a perforated stopper about 1J inches (4.5 centimeters) long and 1J inches (3.8 centimeters) in diameter, with another tube, which has two arms, one, D, to receive the delivery-tube from the pump, the other to 'connect by a rubber tube with the bulb, F, of 200 c.c. capacity, for the supply of potash solution. The burette is enclosed in a water- jacket of about 1} inches (4.5 centimeters) external diameter. Its lower end is closed by the rubber stopper that connects the burette with the two-armed tube below. The upper end of the 1 1868: Zeitsch. anal Chem., 7, 430. See also ibid., 1881: 20, 257. 222 ELEMENTA RY ANAL YSIS. H jacket is closed by a thin rubber disk slit radially and having four perforations : one in the centre admitting the neck of the burette, and three others near the circumference. Through one of the latter a glass tube, L, bent as in the fig- ure, reaches to the bottom of the jacket, another short tube passes through the disk (these tubes conveying water to and from the jacket), and the third hole supports the thermometer. The azotometer is held upright and firm on a stand by rings fitted with cork wedges around it. The bulb for the potash solution rests in a slotted sliding ring. The air-pump* used by Prof. Johnson is a Sprengel mercury-pump, modified so as to be easily constructed and durable. It is shown in outline, with some parts enlarged, in Fig. 19. Through a rubber stopper wired into the nozzle of the mercury reservoir, A, passes a glass tube, B, 4 inches (10.2 centimeters) long, and this connects by a stout rubber tube, C, with the straight tube, D, 3 feet (91.4 centimeters) long. The stout rubber tube, E, 6 inches (15.2 centimeters) long, connects D with a straight glass tube, F, of about the same length as D. G is a piece of combustion tubing, 1J inches (3.8 centimeters) long, closed below by a doubly perforated soft rubber stopper admitting the tubes F and H, and above by the singly perfo- rated rubber stopper into which the tube I is fitted. The tube H has a length of 45 inches (114.3 centimeters). At the bottom it is con- nected by a fine black rubber tube (previously soaked in melted tallow) with a straight tube of 3 inches (about 7 centimeters), and this again in the same way with the tube K, of 7 inches (about 18 centimeters) length. The tubes H and K should have an internal diameter of 1.5 millimeters, F may be 2 millimeters, and D still larger. For H and F may be used slender Bohemian glass tubes of 4 millimeters external diameter. Their elasticity compensates for their slenderness. If heavy barometer tubes be used the stoppers and G must be of correspondingly larger dimensions. The joints at G must 2 A mercurial pump for nitrogen is also figured and described by DABNEY, 1885: Am. Chem. Jour., 6, 236. Kg, 19 ESTIMA TION OF NITROGEN. 223 be made with the greatest care. It is best to insert the lower stopper for half its length into Gr, and F and H should fit so snugly as to be inserted with effort when oiled. The tube I must be of stout glass, about a centimeter (0.4 inch) in diameter, and drawn at both ends to a gradual taper, the outer end bent to connect with the combustion- tube, the inner end when oiled turned into a perforation of about 0.5 centimeter (0.2 inch) in the upper stopper. The joints entering G are the only ones having to resist pressure into vacuum, and they must be made with the utmost care. If not secure without, they are to be trapped with glycerine. To do this pass F and H through a stopper of \ inch (or 13 millimeters) greater diameter than G and placed below it, when, before inserting I, a jacket-tube 4 inches (10 centimeters) long is fitted upon this stopper, surrounding G. After I is in- serted the trap is ready, to be filled with concentrated glycerine, which is preserved from dilution by adding a stopper to the outer tube, around I, split in halves for adjustment. The two rubber tubes are both provided with efficient screw-clamps to govern the flow of mercury. The tubes D, F, H, and I are secured by cork clamps and wires, or otherwise, to an upright plank, which is framed below into a heavy horizontal wooden foot on which rests the mercury-trough. The plank carries above a horizontal shelf for the support of the reservoir, A, the neck of which rests in a perforation in the shelf. At the fastenings of the tubes upon the upright support thick rubber tubes are interposed as elastic rests. The rubber tube joints should be wound with waxed silk. A glass funnel is used in A to prevent spattering of the mercury. I I I | : JO | LJ { f JKCIOs j MIXTURE 'RINSINGS! Cu. |CuOj C0a ! ASBESTOS | fi cm.! 30 cm. S cm. ; 12 cm.J5cm.j3cmi 10cm. I '. 20 The combustion -tube is charged as follows : Of the potas- sium chlorate from 3 to 4 grams, according to the amount of carbon to be burned, are placed in the tail of the tube, Fig. 20, followed by a plug of ignited asbestos just at the bend. Of the substance under analysis 0.6 to 0.8 gram, from the weighing- tube, is well mixed in a mortar (previously rinsed with the copper oxide) with dry (recently ignited) oxide of copper enough to fill 11 or 12 inches (28-30 centimeters) of the tube, and the mixture introduced through a funnel. The rinsings of the mor- 224 ELEMENTAR Y ANAL YSIS. tar with oxide of copper are added to fill about 3 inches (Y.6 centimeters) of the tube, and a second asbestos plug placed. On this is placed the reduced copper for 4 or 5 inches (10 or 12 centimeters), then a third asbestos plug, then 2 inches (5 centi- meters) of the copper oxide, and a fourth plug of asbestos, fol- lowed by 0.8 to 1.0 gram of the sodium bicarbonate. 1 The re- maining space is loosely filled with asbestos to take the water of combustion and prevent it from flowing back upon the heated glass. The anterior part of the tube is wound with copper foil, leaving the rear of the metallic copper visible. The filled com- bustion-tube is placed in the furnace, on a level with the tube, I, of the pump (Fig. 19), and carefully connected with the latter by a close-fitting rubber stopper moistened with glycerine. The azotometer is prepared and tested as follows : The bottom is fill- ed with mercury to about the level indicated by the dotted line Gr (Fig. 18). The arm D is securely closed by a rubber stopper. The stop-cock H is greased, the plug inserted, and the cock left open. The potash solution is poured into F until A is nearly full, and some solution remains in the bulb F, which is now raised care- fully in one hand, while the other hand is upon the stop-cock H. When the solution has risen in A very nearly to the glass cock, the latter is closed, avoiding contact of the alkali with the ground glass bearings, when the bulb is replaced in the ring arid lowered as far as may be. If the level of the solution in the azotometer does not fall in 10 or 15 minutes, it is tight. The pump is set in operation by putting its delivery-tube K in a trough of mercury, supplying the reservoir, A, with at least 500 c.c. of mercury, and cautiously opening the clamps C and E. If the mercury does not start at once, repeatedly pinch the rub- ber at E. It should flow nearly as fast as it can be discharged at K, and without filling the cylinder G. A complete exhaustion CLADDING (1882: Am. Chem. Jour., 4, 45) dispenses with chlorate of pot- ash, and puts about 0.6 gram bicarbonate of soda in the tail of the tube (1). The space 2 is filled with about two inches of ignited asbestos. The substance at 3 is mixed with copper oxide, as fine as sea-sand, without dust. At space 4 is another 0.6 gram of the bicarbonate; then is placed a layer of copper shot, and again a layer of coarsely granulated copper oxide (6). The analysis is begun by drawing the potash solution nearly to the top of the azotometer, then turning up lamps under 6, and at the same time starting the pump. When a perfect vacuum has been obtained and the copper oxide (6) is red hot, the lamp just beyond 1 is turned up, and a gentle heat, just sufficient to drive off the carbon dioxide from, it and not to heat space 3, is applied. When the tube is full of carbon dioxide this lamp is turned off and the tube again exhausted. By this process of washing out the tube several tenths of a c.c. of additional gas are obtained and almost the last traces of air removed. On running the heat back the bicarbonate at 4 gives off carbon dioxide, and refills the tube before the combustion of the substance at 3 begins. ESTIMA TION OF NITROGEN. 225 of the combustion-tube can generally be obtained in 5 to 10 minutes' working of the pump. If the mercury becomes ex- pended before the desired exhaustion is obtained, the clamp C is closed and the mercury returned to A. Complete exhaustion is denoted by a clanking or rattling sound of the falling mercury, and a half a minute after this is heard the clamp C may be closed. If the mercury column in H remains stationary for some minutes, the connections are tight. The mercury trough is closed and the tube K placed in a capsule. Before connecting the azoto- meter, heat is applied to the part of the combustion-tube contain- ing the bicarbonate of sodium. Water-vapor and carbon dioxide are evolved, filling the vacuum in the pump and displacing the mercury in the tube H. The azotometer is placed at hand, its bulb F is taken from the ring and supported in a box near the level of the tube D, the stopper of which is now removed with- out greatly changing the level of the mercury (G). The tube D is filled half full or more with water. As soon as the mercury has fully escaped from the pump-tube K, this is inserted in the azotorneter-tube D. A few bubbles are allowed to escape through the water, and then the tube K is passed down so that the gas escaping from the pump enters the azotometer. It will facilitate the delivery of the gas if the extremity of the pump- tube just touches the inside of the azotometer-tube, as near as possible to the surface of the mercury. The carbon dioxide is absorbed in passing through the caustic potash solution, and no permanent gas should be obtained. In spite of all precautions very minute bubbles of permanent gas will occasionally ascend, but, as will be seen on observing the amount of potash solu- tion so displaced, the error thereby occasioned is extremely small. In the combustion the anterior cupric oxide is first heated to full redness, and then the metallic copper. Then the com- bustion of the substance is steadily carried on, so that the flow of gas into the azotometer is about one bubble a second, or a little faster. When the horizontal part of the tube has all been heated, and the evolution of gas has nearly ceased, the potas- sium chlorate is heated so as to boil vigorously with genera- tion of oxygen. Any remaining carbon of the substance now burns rapidly, and the reduced copper oxide is promptly reox- idized. When the layer of metallic copper in the anterior part of the tube begins to be oxidized, the generation of oxygen is stopped and the heat lowered all along the tube, keeping the metallic copper still at faint red heat. After a few minutes now the pump is started, slowly at first, having some vessel 226 ELEMENTARY ANAL YSIS. under the azotometer-tube D to receive the mercury. A few minutes' pumping suffices to clear the tube, full exhaustion be- ing indicated as stated on p. 225. 1 The azotometer is now removed from the pump, the azoto- meter-tube D is closed by its rubber stopper, the bulb (F) is raised in its ring to such a height that the potash solution in it is nearly on a level with that in the burette, the iilling-tube L is connected with water-supply, a thermometer is inserted in the top of the water-jacket, and the water allowed to run until the temperature and the volume of the gas are constant. The level of the solution in the bulb is now accurately adjusted to that in the burette, and the temperature and the volume of the gas are read, as also the height of the barometer. When 50 per cent, potash solution is used no correction for tension of aqueous vapor is used by Prof. Johnson, following the authority of ScHiFF. 3 The volume read off is reduced to volume at C. by divid- ing by 1 -f- (degrees temperature C. observed X 0.003665). That c.c of observed volume is, - -TTT^ = c.c. volume at C. ' 1 + (observed temp. 0. X 0.003665) The volume at observed barometric pressure is reduced to volume at 760 millimeters barometric pressure by the (inverse) proportion, 760 : mm. of observed pressure :: c.c. observed vol. : x = c.c. at 760 mm. At C., and 760 mm. bar., 1000 c.c. of (dry) nitrogen weigh 1.25616 grains. The corrections, therefore, may be stated : mm. bar. X f 1 ^ a * T tem P erature ' (1 +0.00867 760 The value of this fraction is given in a table for T to 30, by J. T. BROWN: Jour. Chem. Soc., [2], 3, 211; Watts' s Diet. Chem., vi. 147. Correction for temperature, pressure, and water- vapor tension is made by the formula : 1 See GLADDING, under p. 224. 9 HUGO SCHIFF, 1868: Zeitsch. anal. Chem., 7, 432. This author found in several determinations that air dried by passing through a 50 per cent, potash solution, at 24 C., still contained only 108 to 113 milligrams water in 19 liters. This would give to nitrogen a reading about 0.007 of its volume too high. His determinations of nitrogen, by his procedure in the absolute method, were uniformly a little too low, thus: 12.9 instead of 13.2; 31.4 instead of 31.8: 9.0 to 9.1 instead of 9.1; 3.8 to 3.9 instead of 3.9. The deficiency he ascribed to retention of traces of nitrogen oxides. And the author advises to neglect the correction for aqueous vapor, in compensation for the margin of loss. ESTIMA TIOX OF NITROGEN. 227 P = 0.0012562 X V X . 036T T) Wherein P = the grams weight of the nitrogen measured. V = c.c. of observed volume. T = temperature of the azotometer-jacket in de- grees C. B = millimeters of barometric reading. f = tension of water-vapor, at T, found in milli- meters. Of the tables convenient for consultation, to shorten calcula- tions for nitrogen, are those of BATTLE and DANCY, for use in Analysis of Commercial Fertilizers, 1885 : North Carolina Ex- periment Station, Raleigh, N. C. Also, for general uses, KOHL- MAN und FREKICHS, " Rechentafeln," 1882 : Leipzig. The correction for water-vapor tension is purposely neg- lected by some chemists, on the ground (already mentioned) that strong potash solution leaves the gas nearly dry. 1 On the other hand, the results by Johnson's procedure in absolute method for nitrogen are more apt to be over than under the true quantity (see the citation from DABNEY, under Ruffle's Method). "When the correction is required it is made as follows : Consult a table of Tension of aqueous vapor at various temperatures (this tension being irrespective of pressure), and find the tension, in height of mercury, for the observed temperature. Subtract this tension from the barometer reading in the operation in hand, as in the formula above. Method of Maxwell Simpson (1855). Combustion by a mix- ture of copper oxide with mercury oxide, the tube having been cleaned of air by a current of carbon dioxide liberated by heat- ing a carbonate. The excess of oxygen is taken up by a good quantity of heated metallic copper in the combustion-tube; the carbon dioxide by potash solution in a receiver ; and the nitrogen is measured over mercury for the calculation of its weight. The mercuric oxide is to be prepared by precipitation with fixed al- kali, washing with water and then with dilute phosphoric acid, and drying at 100 C. The combustion-tube, about 80 centi- meters (31.5 inches) long, is closed in a rounded end by fusion. 1 Owing to the fact that the strength of the potash solution varies, and the water-vapor tension is therefore uncertain, GALTERMAN (1885) collects the ni- trogen over potash solution in a non-calibrated tube, thence transferring it to a measuring-tube over distilled water. The full tension of the water-vapor is deducted. 228 ELEMENTARY ANALYSIS. A mixture of 12 grams of manganese carbonate or of magnesite, previously dried at 100 C.j with 2 grams of the mercuric oxide, is introduced into the tube. A plug of recently ignited as- bestos is inserted, pushing it down to within 3 centimeters (about 1 inch) of the mixture, and next is added 1 gram of the mer- curic oxide. Of the substance under analysis about 0.6 gram is taken, from a weigh ing- tube, for intermixture in a mortar with 45 times its weight of a prepared mixture of 4 parts of finely pow- dered and recently ignited copper oxide, with 5 parts of the dried mercuric oxide. The whole is transferred to the combustion- tube, the mortar is rinsed with some more of the mixed oxides, and the rinsings added. A second plug of ignited asbestos is pushed down to within about 30 centimeters (near 12 inches) of the first, leaving the mixture of oxides loose ; a layer of 6 to 9 centimeters (2J-3J inches) of the copper oxide is added and a third plug of asbestos placed ; and lastly a layer of as much as 20 centimeters (near 8 inches) of metallic copper, prepared by reducing granular copper oxide in a stream of hydrogen at low temperature (or in a stream of carbon monoxide). The com- bustion-tube is now drawn out and turned down, and connect- ed by a section of rubber tubing with a delivery-tube adapted to reach beneath the surface of mer- cury in the trough. The combustion-tube is tapped on the table to form a channel for the escape of the gases, and placed in the furnace. A receiver is provided, as shown, with the trough of mercury, in Fig. 21. The receiver has about 200 c.c. capacity ; the glass stop-cock should enable it to hold mercury when filled with it and set up in place ; a delivery-tube is firmly connected with its neck, and it is tubulated on the side near its base. This tubule carries an upright filling-tube, with contrac- tion near the tubule. It is filled with mercury, placed in the trough with the tubule under the mercury, and about 20 c.c. of strong solution of potassium hydroxide passed into it. A meas- uring-tube for the nitrogen gas is represented in Fig. 22. But instead of both the receiver and measuring-tube here described, the azotometer figured on p. 221 may be used. About half of the carbonate in the posterior end of the com- ESTIMA TION OF NITROGEN. 229 bustion-tube is heated, so that the air is driven out by a current of carbon dioxide ; and at the same time a part of the tube oc- cupied by the metallic copper and the copper oxide is heated. The escaping gas is tested for air, from time to time, by receiv- ing a few bubbles in an inverted test-tube containing solution of potash ; and when the bubbles are completely taken up by the solution, and the anterior part of the tube is well heated, the delivery-tube from the combustion is inserted in the lateral tubule of the receiver. The substance in mixture with the oxides is now gradually heated, beginning next the clear copper oxide, until the whole tube, except that occupied by carbonate in the rear, has been at full heat, and no further delivery of gas is ob- served. Next, the remain- der of the carbonate is heated, so as to sweep out the nitrogen remain- ing in the tube. The delivery-tube is now withdrawn from the re- ceiver, which is left for an hour for the absorp- tion of the last traces of carbon dioxide. The nitrogen gas is transferred to the measur- ing-tube, Fig. 22. The stopper inserted into the lateral tubule of the receiver is moistened with mercuric chloride solution to prevent its carrying in air. A drop of water is placed in the measuring-tube before it is filled with mercury and inverted in the cistern. The stop-cock in the neck of the receiver is care- fully governed to obtain a very gradual delivery of the gas, and is closed each time that the mercury is poured into the filling- tube, below the contraction in which the mercury is not permit- ted to fall in the beginning of the transfer. Close the stop-cock as soon as it is reached by the potash solution, leaving the ni trogen in the delivery-tube to compensate for the air it contained to begin with. For calculation of weight from volume, with corrections for temperature and pressure, see p. 226. A VERY SIMPLE METHOD FOR ABSOLUTE DETERMINATION OF NITROGEN, when carefully conducted, will give good results. 230 ELEMENTAR Y ANAL YSIS. An operation as follows, with copper oxide as the sole supply of oxygen, with SchifFs azotometer, and without a pump, will give true results, though requiring more time than the method of Johnson or that of Simpson. The copper oxide is dried bj ignition in a current of dry air in a combustion-tube with bayo- net-end. In a combustion-tube of good length, closed (with round end) at the rear, a layer of manganese or magnesium carbonate is placed first, as stated on p. 228, then a plug of as- bestos, then a short layer of copper oxide, then the substance mixed with copper oxide, mixing in a mortar or in the tube. About two-thirds of the tube should remain for the layers of copper oxide and metallic copper. The latter may be a roll of ignited copper gauze or a layer of reduced granular oxide, and should be 5 to 8 inches long. Anterior to this may be, as pro- posed by Professor Johnson, a short layer of copper oxide to oxidize any occluded hydrogen. In the combustion the air is first expelled by liberating car- bon dioxide from a part of the carbonate in the rear ; the ante- rior layers of metallic copper and copper oxide are kept at full red heat ; the substance is burned very slowly, and much time is taken in oxidizing the last of the carbonaceous residue ; and finally the tube is swept out by ignition of the remaining car- bonate in the posterior end. The gases from the tube are re- ceived directly into a Schiff's azotometer, over strong potash solution. In measuring the nitrogen, the room and apparatus being of uniform temperature, a thermometric reading is ob- tained. ESTIMATION OF ORGANIC NITROGEN BY ITS CONVERSION INTO Ammonia. The SodctrLime process of Varentrapp and Will. The nitrogen of nitrates is not included in this estimation. The substance is heated in a combustion-tube in mixture with soda- lime, the products being carried through a layer of red-hot soda-linle of at least half the length of the tube, and received in a solution of acid. The ammonia remaining in the tube after the combustion is swept out by burning a short layer of oxalic acid in the rear, also by aspiration. If the substance be rich in nitro- gen it is diluted with cane-sugar. The gaseous ammonia from the combustion-tube is received in a known volume of a standard solution of oxalic or sulphuric acid, which is afterward titrated (PELIGOT'S modification) ; or is received in hydrochloric acid for gravimetric estimation with platinic chloride. Using Peligot's modification, Prof. S. W. JOHNSON found 1 that, with various '1879: Am. Chem. Jour., i, 75; 1872: Am. Chemist, 3, 161. ESTIMA TION OF NITROGEN. 231 substances, under a series of determinations, " the soda-iime pro- cess is, to say the least, equal in accuracy with the absolute determination," by volume of free nitrogen. At bright red heat, with soda-lime, ammonia is not decomposed. A combustion-tube of 14 to 30 inches (35 to 75 centimeters) length, and near \ inch (10 to 12 millimeters) width, is sealed round at one end (Fig. 23). The 7&?\Qi\mQyQ^&' gas-furnace is the most convenient. The best lulled U- tule is that shown in the figure. The acid is of about normal strength, titrated with an alkali solution of about half -normal, the latter being exactly valued with a standard acid solution prepared with care. Prof. S. W. Johnson uses standardized hydrochloric acid and standard solution of ammonia, and titrates with cochineal tincture as an indicator. The same indicator should be used in all titrations ; and if me acid solution become colored from the combustion, litmus tincture is not applicable. Litmus-papers, blue and red, serve vejy well. The soda-lime, preferably granulated, otherwise coarsely^ powdered, is heated to remove all moisture, which is strictly excluded until the article is used. It may be used warm if the sil'bstance is stable enough to suffer no change therefrom. Oxalic add should be heated on the water-bath to remove all water of crystallization. Asbestos, recently ignited, is required. In the charging of the combustion-tube a layer of about 1 inches (3 centimeters) of the dried oxalic acid is intro- duced into the rear of the tube, followed by about the same length of soda-lime. The substance under analysis is added from the weighing-tube, in quantity about 0.5 gram, to some of the soda-lime in a mortar (previously rinsed with the soda-lime), and a mixture made which, with the rinsings of the mortar, will fill the tube to a point from two-fifths to one-half its length from the closed end. Or the mixture of the substance with the soda-lime is made in the tube by means of a stirring-wire (Fig. 8), so as to form a layer of near the length just stated. In either case, if the substance be very rich in nitrogen, about an equal quantity of dried cane-sugar may be taken with it in the mixture. The remainder of the tube is filled with the soda-lime to within about 2 inches (5 centimeters) of the rubber stopper, placing a loosely porous plug of the asbestos, nearly an inch (or 2 centimeters) in length, as a secure guard against the carrying forward of alkaline 232 ELEMENTAR Y ANAL YSIS. dust or spray, and leaving a free space next the stopper. A shield may be put over the end of the tube (Fig. 16). The U-tube is filled and connected as shown in Fig. 23. The more that moisture has been excluded from the soda-lime, the easier will be the combustion. But the use of warm soda- lime in intermixture with the substance must not be adopted without assurance that no traces of ammonia are generated in such mixture. If the soda-lime be well granulated, or even coarsely powdered, with fine particles sifted out, it is better not to triturate in making the mixture of the substance, and to do without a channel formed by tapping the horizontal tube on the table, favoring the more intimate contact of empyreumatic gases with the hot soda-lime. But if there are layers of line powder in the tube, a channel must be provided. In the combustion the layer of unmixed soda-lime is first heated, beginning at the anterior end, and increasing and extend- ing the heat at such a moderate rate that the air-bubbles shall not pass out faster than about two to each second. The heat at the anterior end is so graduated as to prevent condensation of water- vapor in the tube, and not to soften the rubber stopper. When the mixture of substance is reached the layer of clear soda-lime must be at full red heat, and so preserved while the flames are advanced backward more gradually than before, delivering only about one bubble every second. The carbonized substance is at last burned out with a full red heat, and when the delivery of gas has nearly or quite ceased the oxalic acid is very gradually heated, so that the carbon dioxide shall not be tumultuously evolved. The carbon dioxide is generated only long enough to sweep out the combustion-tube, when the U-tube may be de- tached. The acid liquid should be as little colored and empy- reumatic as possible. The anterior end of the combustion-tube, in the space in front of the asbestos plug, should not change moistened red litmus-paper. In titrating the acid for the amount of ammonia it has re- ceived, the volumetric alkali is added from the burette directly to the U-tube until the neutral point is very nearly obtained, with litmus-papers or other indicator, not phenol-phthalein. The acid is now transferred to a beaker, with ' very little rinsing-water, and the titration completed. The value of the alkali solution is found by a volumetric acid of absolute stand- ard. 17 : 14 :: quantity of ammonia : x = quantity of nitrogen. Combustion-tubes with the posterior end drawn out are some- times used, and the residual ammonia obtained by aspiration, or by sending through a current of carbon dioxide. ESTIMATION OF NITROGEN. 235 The gravimetric determination of the ammonia, as ammo- nium platinic chloride, is done by the ordinary method, as found in works on inorganic analysis, washing the precipitate with alco- hol or ether- alcohol, and igniting in a weighed crucible. 194.4 parts of Pt represent 14 parts of N". Combustion with soda-lime in an iron tube may be done with good results, 1 as the writer has verified. The tube should be about a third longer, and a little wider, than a glass tube for the same combustion. Special precaution is necessary to avoid burn- ing or melting the stoppers. Combustion with soda-lime, sulphur, and thiosulphate. RUFFLE'S METHOD, 1881. Reduction by a powerful deoxidizer in presence of a strong alkali. Obtains the nitrogen of organic, ammoniacal, and nitric combinations. Carried out in the same way as the Yarentrapp -Will method in Peligot's modification. The method has been well sustained. DABNEY (1885, already cited) found this method, in application to fertilizers containing small amounts of nitrogen, to give results as close as those by Johnson's process for free nitrogen, the latter method giving often a little too high, the former a little too low figures for the nitrogen. Greater precautions are required for bodies rich in nitrogen. Details are presented in the Official Methods of the Association of Agricultural Chemists for 1886-7, Bulletin No. 12, Department of Agriculture, Washington, 1886. RELATIVE DETERMINATION OF THE NITROGEN AND CARBON. Applicable when the proportional quantity of nitrogen is not small, or not less than 1ST to 4 C = 14 of nitrogen to 48 of car- bon. The substance is burned, with copper oxide, and the products passed over hot metallic copper, in a combustion- tube, so as to deliver in a graduated tube the nitrogen and the carbon dioxide. After taking the volume measure of the gases the carbon dioxide is taken up by alkali and measurement taken again. Methods of Liebig, Bunsen, and Gottlieb are employed. THE DETERMINATION OF CARBON, HYDROGEN, AND NITROGEN, in one operation, is described by C. G. WHEELER, 1866 : Am. Jour. Sci., [2], 41, 33. Also by W. HEMPEL, 1878 : Zeitseh. anal. Chem., 17, 409; Jour. Chem. Soc., 36, 278. Recently by P. JANNISCH and Y. MEYER, 1886: Ber. d. chem. Gesel., 19, 949 (preliminary notice). also JOHNSON, 1879: Am, Chem. Jour., i, 82. 234 ELEMENTAR Y ANAL YSIS. THE DIRECT ESTIMATION OF OXYGEN lias been reported upon as follows : BAUMHAUER, 1866 ; MAUMENE, 1862 ; MITSCHERLICH, 1867, 1868; LADENBURG, 1865; CRETIER, 1874. ESTIMATION OF NITROGEN BY COMBUSTION IN THE MOIST WAY. The well-known process published by Prof. WANKLYN in 1877 depends on the conversion of the nitrogen of organic compounds into ammonia by the action of permanganate in a very dilute solution of alkaline reaction, the ammonia already contained in the substance being previously distilled off. Its value, in water analysis, is relative rather than absolute, and depends upon its applicability to nitrogenous organic compounds in an extremely dilute solution, so that the changes likely to occur in a concen- tration of the water are avoided. For the analysis of pure ni- trogenous compounds various plans of moist combustion have been proposed of late years. Of these the following method has received general commendation from chemists who have reported trials of it a method in which oxidation by adding dry perman- ganate to a concentrated acid solution is preceded by the altera- tive action of hot sulphuric acid of full strength : Moist Method of KJELDAHL/ For bodies moderately rich in nitrogen 0.250 gram is taken ; for bodies with only about 1.5# of nitrogen 0.7 gram is taken. The substance is placed in a boil- ing-flask of about 100 c.c. capacity, with a long and narrow neck, and of glass capable of resisting the strongest acids. The flask is placed upon asbestos cloth or copper gauze over a lamp supplying a strong heat, 10 c.c. of pure sulphuric acid of full strength is added, and digestion instituted (under a hood) at a temperature .only a little below the boiling point of the sulphuric acid. Sulphurous acid vapors escape. To prevent loss by spirting, the flask is somewhat inclined during the effervescence. After the liquid comes to rest the digestion is continued (still near the boiling point, as shown by occasional bumping) until the liquid becomes gradually of light color, and finally entirely clear. To 1 J. KJELDAHL, Carlsberg Laboratory of Copenhagen, 1883: Zeitsch. anal Chem., 22, 366; Ohem. News, 48, 101; Am. Chem. Jour., 5, 456. FRESENIUS* 1884: Zeitsch anal. Chem., 23, 53. CZECZETKA, 1886: Monatsch Chem , 6 68; Jour. Chem. Soc., 48, 688. WILFARTH, 1885: Chem. Cent., 1885, 17; Jour. them. Soc. t 48, 837. BOSSHARU, 1886: Zeitsch. anal. Chem., 24, 199; Jour. Chem. Soc., 48, 837. 0. ARNOLD, 1886: Archiv d. Pharm.. [3], 23, 177; Jour. Chem. Soc 48, 688. Details are defined in the " Official Methods of the Association of Agricultural Chemists," Department of Agriculture, Bulletin No. 12, Washington, 1886. The use of phenolsu! phonic acid is introduced into the process by JODLBAUEE. 1886: Chem. Cent., p. 433; Jour. Chem. Soc., 50, ESTIMA TION OF NITROGEN. 235 hasten this result a little fuming sulphuric acid or phosphoric anhydride is added. With these additions a digestion of about two hours is usually sufficient. But at 100 to 150 C. the for- mation of ammonia is imperfect and the object not attained. The lamp is now removed, and, while the liquid is hot, finely pulverized potassium permanganate is carefully added, either in very small portions or in a very fine stream, which may be car- ried through a delivery-tube. The reaction is violent, even ac- companied by small flames, and it is made as gradually as it can be without interrupting it. When the oxidation is complete a green color appears, and the addition of the permanganate is dis- continued. The liquid may now be warmed for a few minutes, but not on any account strongly heated. The liquid is cooled, and diluted with water, when the green color changes to brown. When again cool the liquid is introduced into a distillatory apparatus, the generating flask holding about f liter, and con- nected with an upward-sloping top-piece to prevent liquid being carried over by spirting, and through the condenser into a re- ceiver containing an accurately measured quantity of acid of known strength. About 40 c.c. of solution of sodium hydrate of sp. gr. 1.30 are quickly introduced into the distilling flask. [A Welter's safety tube may be provided for this purpose.] And to prevent bumping a little metallic zinc is introduced, the hydrogen from which secures an even action. The completed distillate is titrated for the ammonia it has received (as in the estimation of Varentrapp and Will). Kjeldahl found his method inapplicable to -certain alkaloids, cyanides, volatile acids, and nitrogen oxides. It reduces nitrates in presence of organic matter to ammonia, but incompletely (com- pare WARINGTON, 1885 : CJiem. News, 52, 162). Upon the Determination of Total Nitrogen, organic, am- moniacal, and nitrous, see BULLETIN No. 12, Chemical Division, DEPARTMENT OF AGRICULTURE, Washington, 1886, pp. 34, 52. Also, GERMAN MANURE MANUFACTURERS' ASSOCIATION, 1884: H. H. B. Shepherd, translator. Also, HOUZEAU, 1885. RUF- FLE'S method to this effect is referred to on p. 233. BODIES CONTAINING SULPHUR, in estimation of carbon and hydrogen, are subjected to combustion with lead chromate in- stead of copper oxide, and the front of the column of lead chro- mate is not heated to full redness. WHEN CHLORINE, BROMINE, or IODINE is present, in combustion to estimate carbon and hy- drogen, a coil of silver wire is placed in the front of the combus- tion-tube to retain the halogens, which otherwise may interfere 236 ELEMENTARY ANALYSIS. with the result. Chlorine forms cuprous chloride, which will condense in the calcium chloride tube. Copper holds chlorine but imperfectly, and the same is true of lead. 4 THE ESTIMATION OF SULPHUR, in organic analysis of com- pounds not volatile, may be done by fusing with potassium hy- drate and nitrate, in a silver dish, until the^mass will be white on cooling. The mass is dissolved in water, acidified by nitric acid, and the quantity of sulphuric acid determined by precipita- tion with barium chloride in the manner required in quantitative work. Volatile compounds can be oxidized with a mixture of sodium carbonate and potassium nitrate in a combustion-tube. Potassium dichromate is also employed as an oxidizing agent in the same way. CHLORINE, BROMINE, and IODINE are estimated by igniting the substance with an excess of pure quicklime, in a narrow combus- tion-tube. The tube is filled with the lime mixed with the sub- stance, followed by a short column of lime alone, and a channel made by tapping the tube on the table. After the ignition the contents of the tube, when cold, are carefully transferred to a flask containing water, and treated with dilute nitric acid, rinsing the tube with the water and then with the acid. The solution is filtered, the residue and filter washed, and the halogens precipi- tated by silver nitrate solution. With iodine it is better to ex- haust first with water, and add silver nitrate solution to the filtrate, then treat the residue with dilute nitric acid and add the acidulous filtrate to the one containing the silver. By this means the liberation of iodine by action of nitric acid is avoided. The silver precipitate is treated as in ordinary quantitative work for the halogens. ESTIMATION OF SULPHUR OR OF HALOGENS is effected by the method of CARIUS 1 From 0.15 to 0.40 gram of the substance is treated with a calculated quantity of nitric acid sufficient to furnish 4 times the required amount of available oxygen, or of acid of sp. gr. from 20 to 60 times the weight of the substance. The digestion is done in a closed tube, sealed, at 120 to 140 C., for some hours. For estimation of chlorine, silver nitrate is added with the nitric acid before digesting. Details may be found in the original papers and in manuals of quantitative analysis. 1 1860-65: Ann. Chem. Phar., 116, 11; 136, 129; Zeitsch. anal. Chem., i 217,240; 4, 451; 10, 103. DEDUCTION OF CHEMICAL FORMULAE. 237 DEDUCTION OF CHEMICAL FORMULAE. In the first place, the molecular weight of the substance is to be ascertained, if possible. (1) If the compound be sufficiently vaporizable its Vapor Density 1 is to be determined and accepted as evidence of the molecular weight. With the weight of air as the unit of gravi- ties, vapor density X 28. 86 = molecular weight. With hydrogen as the unit, vapor density X 2 molecular weight. (2) If the substance have a definite capacity of combining, as a base or an acid, its combining number can be determined by its proportions in formation of salts. If an acid, it is needful to ascertain whether it be monobasic, bibasic, or tribasic in its capacities of combination. Certain classes of bases are subject to the corresponding question, whether nionacid or not, but the natural nitrogenous bases are mostly monacid. (3) If the substance be found to hold a definite relation to other substances, as shown by its formation, its decomposition, or by chemical resemblance to members of homologous series, its molecular weight may be inferred from such relations. If now m be the molecular weight of a compound ; p, the percentage of a constituent element ; a?, the combining number of this element ; #, its atomic weight, and y, the number of its atoms in the molecule, 100 : p : : m : x. And x -h a = y. Whether the molecular weight be obtainable or not, an empi- rical statement of atomic numbers can be derived at once from the centesimal figures of the analysis by dividing the percentage number of each element by its atomic weight. The provisional formula so obtained is reduced, by common divisors, to lower terms, and to such terms as best accord with its probable molecu- lar weight, in its apparent classification among compounds of known molecular formulae. Allowances must be made for the real limits of error in analysis, and consideration must be had to the liability of weigh- 1 For ordinary purposes the most ready and satisfactory method of obtain- ing Vapor Density is that of VICTOR and CARL MEYER, 1878: Ber. d. chem. &es., 10, 2253; Zeitsch. final. Chem., 19, 214; "Watts's Diet. Chem," 8, 2094. .See also reports by V. MEYER, 1876-7: Ber. d. chem. Oes., 9, 1216; 10, 2067; 11, 1867; Zeitsch. anal. Chem., 18, 294; 17, 373. HOFM ANN'S Method was given in 1868: Ber. d chem. Ges., i, 198; Zeitsch. anal. Chem.,B, 83. BUNSEN'S Method, 1867: Ann. Chem. Phar., 141, 273: Zeitsch. anal. Chem., 6, 1. Method of TROOST and DEVILLE, 1860: Ann. Chim. Phys., [3], 58, 257. A good summary of the literature of vapor-density determination is given in "Roscoe and Schorlemmer's Chemistry," vol. 3, part 1, p. 84 and after. Also see " Beilstein's Organische Chemie," 2d ed., p. 17. 238 FA TS AND OILS. able impurities, including moisture, in the article taken for com- bustion. Probable limits of error are represented in general by a comparison of published results of analyses by authorities of credit, and, more definitely, by the experience of the analyst himself with substances of known composition. Even in empirical formulae the well-known law of conjugate atomic numbers of carbon compounds should be respected, namely, the numbers of the atoms of uneven valence (the perissads, H and N) should together make an even number. Thus in the molecule of morphine, with 'N 1 we have H 19 ; in the molecule of quinine, with JsT 2 we have H 2 4- That is, in ordinary non-nitro- genous organic molecules, those containing C, H, and O, or those containing C and H, there is always an even number of atoms of hydrogen. But in nitrogenous molecules (of C, H, IS", O, or C, H, N) the atomic number of hydrogen is even only when nitrogen presents an even atomic number. The law applies to haloid elements and to phosphorus, when these elements of un- even valence are present. The low atomic weight of hydrogen gives low centesimal dif- ferences for one atom of this element, so that its atomic number is taken as the number which, under the rule, lies nearest the atomic number calculated from centesimals. The establishment of a rational formula for a compound is a work of investigation, both synthetic and analytic, as obtained by reactions of formation and of decomposition. It requires studies of all chemical relations, led on by analogies from every point of view. An understanding of the chemical structure of the molecule is gained step by step in the investigation. A de- rived chemical formula can be made "rational" only to the extent that the chemical forces of the constituent elements have been revealed in their proportional power. In the study of isome- rides, for the " position " of atoms in molecules, the atomic posi- tion is to be defined as a mode of statement of the chemical functions of the elements. At the same time it may be said that the evidence gained for relative u position " of atoms in a mole- cule is of the same character as the evidence upon which we predicate the existence of the molecule as a whole. FATS AND OILS. 1 Glycerides, and bodies related thereto 1 A good general summary of the chemistry and technology of the neutral fats is presented in the article "Chemical and Analytical Examination of Fixed Oils," A. H. ALLEN, 1883: Jour. Soc. Ohem. Indus., 2, 49. A compact technical summary of the analytical chemistry of fats is presented in BENE- DIKT'S " Analyse der Fette und'Wachsarten," Berlin, 1886, pp. 296. FATTY SERIES OF ACIDS. 239 either by physical properties and uses or by production, are treat- ed in the following pages under the heads here given : Fatty Series, C n H an 02: Stearic Acid ; Palmitic Acid; Myristic, Laurie, Capric, Caprylic, Caproic acids. Fatty Series, C n H 2n _ 2 2 : Oleic Acid. Ricinoleic Acid ; Linoleic Acid ; Hypogaic and Physetoleic acids. Fat Acids and Fats, Quantitative Determination of : (1) Hehner's num- ber; (2) Reichert's number; (3) Kottstorfer's number, or capacity of saturation; Tables of Hehner's and Kottstorfer's numbers; (4) Iodine number of Hiibl: (5) Mean Molecular Weight; (6) Specific Gravities; (7) Melting and Congealing points; (8) Calculation of Acid and Neutral Fats. Distinctions of Fat Oils by Solubility in Glacial Acetic Acid ; Table. Separation of Mineral Oils from Fats: descriptive list; method by saponi- fication; extraction after saponification, in solution, in dry mass, with Soxlet's apparatus, estimation by Kottstorfer's numbers ; examination of the liquid and solid non-saponifiable bodies. Separation of Fat Acids from Fats. Separation of Resins from Fat Acids. Rosin Oils. Drying and Non-Drying Oils. Linseed Oil; Olive Oil; Cotton-seed Oil ; Castor Oil; Lard; Tallow; Oleo- margarin. Butter: bibliography: constituents; estimation of constituents, of artificial color, or rancidity (acidity); detection of foreign fats by solvents; Scheffer's method; odor test; soap viscosity ; microscopical examinations; butter fat, properties: butter substitutes ; methods of chemical estimation of butter fat- Header'*, Reichert's, Kottstorfer's; interpretation of Hehner's number, of Reichert's, of Kottstorfer's; specific gravity as a means of distinguishing from substitutes; iodine number of Hilbl; scope of butter analysis and forms of certificates, in Massachusetts, in New York, in Pennsylvania, at Agricul- tural Department at Washington; what is a sufficient chemical analysis of butter. FATTY SERIES OF ACIDS, C n H 2n O 2 . The folio wing- named members of the C n Il2nO 2 series are described in this work, and will be found under their respective names. Formic, Acetic, and Yaleric Acids are not constituents of Fats. The others are described in the next following pages. For Butyric Acid see p. 75. Volatile. Formic Acid CH 2 O 2 or H . CO 2 H Acetic Acid C 9 H 4 O 2 " CH 3 .CO 2 H Butyric Acid normal. . C~ 4 H S O 2 " CH 3 CH 2 CH 2 .CO 2 H Yaleric Acid isovaleric. C 5 H 10 O 2 " (CH 3 ) 2 CH CH 2 . CO 2 H Caproic Acid isobutyl- acetic C 6 H 12 2 (CH 3 ) 2 CH(CH 2 ) 2 . CO 2 H Caprylic Acid normal. C 8 H 16 O 2 " CH 3 (CH 2 ) 6 .CO 2 H Capric Acid C 10 H 20 O 2 CH 3 (CH 2 ) 8 .CO 2 H ? Laurie Acid C 12 H 24 O 2 240 FATS AND OILS. Non- Volatile. Mjristic Acid. C 14 H 28 O 2 Palmitic Acid C 16 H 32 O 3 Stearic Acid C 18 H 3 gO 2 STEAEIO ACID. C 18 H 36 O 2 =:284 (monobasic). Found as a normal glyceride in common vegetable and animal fats, in which it is the ordinary constituent of highest melting point. a. Crystallizable from alcohol in white, lustrous tables, or needles; or congealing from a melted portion in crystalline, translucent masses of considerable hardness. It melts at 69.2 C. At about 360 C. it begins to boil with decomposition of a con- siderable part. Under reduced pressure, at 100 millimeters, it boils at 291 C. With superheated steam it distils with but lit- tle decomposition. Its specific gravity as a solid at 11 C. is that of water at the same temperature, but at higher tempera- tures it floats upon water, and the melted acid just above 69.2 C. has the specific gravity of 0.8454. The melting paint can be found, quickly, by immersing the bulb of the thermometer for a moment in the melted stearic acid (free from water), then suspending the coated bulb in the middle of a beaker of water, to which heat is applied, and noting the tem- perature at which the fatty coat melts from the bulb. To purify stearic acid from salts, preparatory to this test, it may be repeat- edly dissolved in alcohol, filtered, and evaporated to dryness. (Further, see Determination of the Melting and Congealing Points of Fatty Bodies, Index.) The normal glyceride, stearin, or " tristearin," C 3 H 5 (C 18 H 35 O 2 ) 3 , is crystallizable, and, when pure, of pearly- white lustre. It melts, according to modification due to previous heat- ing, at a temperature from 55 C. to 71.6 C. Stearin crystal- lized from ether melts at 71.6 C., and then congeals to a crys- talline mass at 70 C. ; but heated only 4 C. above the melting point, it does not then congeal until reduced to the temperature of about 52 C., when it appears as a wax-like mass and will melt^ again at 55 C. A sample of stearin (not entirely pure), melting at 65. 5 C., at this temperature had the specific gravity 0.9245 (BENEDIKT '). The^ metallic stearates are fusible bodies, in some instances crystallizable, more often amorphous, and of plaster- like or soap- like consistence. b . Stearic acid and stearins are odorless and tasteless. 1 "Analyse der Fette," Berlin, 1886. STEARIC ACID. 241 c'. Stearic acid is insoluble in water. It is soluble in about 40 parts of absolute alcohol at ordinary temperatures, moderate- ly soluble in 90$ alcohol when hot, very sparingly when cold. On cooling the hot alcoholic solution abundant crystals are ob- tained. It is readily soluble in ether. The solutions redden lit- mus-paper, and decolor the alkaline phenol-phthalein. At 23 C. it dissolves in 4.5 parts of benzene or in 3.3 parts of carbon di- sulphide. Stearin, the neutral glyceride, is insoluble in water, some- what soluble in boiling alcohol, from which it crystallizes out almost wholly when cold. It dissolves 'in about 200 parts of ether a solubility more sparing than that of the softer neutral fats and dissolves in chloroform, benzene, petroleum benzin, and carbon disulphide. The alkali stearates are somewhat dif- ficultly and imperfectly soluble in water. They dissolve in hot water, with slight turbidity, the solution gelatinizing when cold. On agitating the gelatinized mass with much water in the cold, a turbid mixture is obtained, with formation of difficultly soluble acid stearate along with free alkali. In hot alcohol the alkali stearates are easily soluble, the greater part congealing in the cold, so that only a dilute solution is permanently obtained. The non-alkali metallic stearates are insoluble in water, and for the most part insoluble in alcohol or ether. In some instances, however, they yield free stearic acid to the action of ether. In general they are gradually decomposed by action of water, yield- ing hydrate of the metal and free stearic acid. d. The aqueous solutions of alkali stearates, dilute and turbid, on addition of solution of barium or calcium chloride, or other non-alkali salt, show an abundant precipitate of metallic stearate. An alcoholic solution of stearic acid, with a solution of barium or calcium acetate to which a little alcohol has been added, gives a precipitate of stearate of the metal. The barium precipitate is gelatinous and bulky ; the magnesium precipitate, crystalline and pulverulent. To the action of water these pre- cipitates yield hydrates of barium, etc., while free fat remains behind. The precipitates are to some extent dissolved by boil- ing alcohol, and on cooling the solution crystalline precipitates are obtained. Solutions of alkali stearates are precipitated by addition of dilute acids, the resulting stearic acid appearing in a milky subdivision with curdy clumps. By heating the mixture a clear, oily layer slowly rises, and on cooling solidifies to a crust. Or the precipitate while cold may be filtered out, washed with hot water, drained dry, and dissolved from the filter with hot 242 FA TS AND OILS. alcohol or with cold ether. On evaporation the stearic acid is obtained, crystalline or congealed, as preferred. Also the crude precipitate of stearic acid may be dissolved by shaking out with several portions of ether. e. Separation. Stearic acid is obtained from its glyceride, stearin, by first saponifying with potash, and then decomposing the soap with acid. The saponification is done by boiling gently with alcoholic solution of potassa until a clear solution is ob- tained. Ten parts of the fat are treated with 8 to 10 parts of 70 to 85$ alcohol, and 4 to 6 parts solid potassa. The most of the alcohol is evaporated off, and the cold liquid treated with dilute acid for liberation of the stearic acid, as directed above (d). From non-alkali stearates, acidulating with an acid that does not precipitate the metal, and shaking out with ether, is usually the most expeditious method of separating the stearic acid. From oleic acid stearic acid (with other solid fat acids) is separated T)y insolubility qf lead stearate in ether, as follows. The free fat acids are to be perfectly saponified with potassa or soda ; the neutral solution of the alkali soap, with some alco- hol, is precipitated with lead acetate, and the precipitate washed, dried, and exhausted with ether in repeated portions, when the lead salts of the solid fat acids will be left undissolved, and the lead oleate can be obtained by evaporating the ethereal solution. The details may be governed as follows (]REMEL *). Of the free fat acids 2 to grams are exactly weighed into a flask of 100 to 150 c.c. capacity, treated with about an equal quantity of dry caustic potash and 10 c.c. of alcohol of &5 per cent, strength, on the water-bath, to complete saponification. The mass is di- luted with a little water, neutralized with acetic acid, using phenol-phthalein as an indicator, the alcohol evaporated off on the water-bath, the residue dissolved in 80 c.c. hot water, and the liquid precipitated with lead acetate solution. When cold the free precipitate is poured upon a filter of 10 cm. (near 4 inches) diameter, and the whole precipitate is washed several times with hot water. The precipitate adhering to the flask is melted on the water-bath, cooled, drained, and dried at a gentle heat, as is also the precipitate in the filter. The contents of the flask are now treated with ether, poured through the same filter, in repeated portions, until the whole precipitate is exhausted of ether-soluble substance. On vaporizing the ether in the filter the lead stearate can be detached, and added to that in the 1 Consult also MUTER : Analyst, 2, 73. STEARIC ACID. 243 flask, where the whole is treated with diluted hydrochloric acid, and exhausted with ether. The liltered ethereal solution is eva- porated in a tared beaker and the residue weighed as stearic acid (including all solid fat acids). For the oleic acid (the total of liquid fat acids) the ethereal solution of lead salt is evaporated to dry ness, and the residue treated with diluted hydrochloric acid and then with ether, as directed for the solid fat acids. Stearic acid (with palmitic acid) is separated from oleic acid by tJie solvent action of a mixture of alcohol and glacial acetic acid (DAVID, 1878 1 ). In the proportion of 300 c.c. of alcohol of 95 per cent, strength with 220 c.c. of glacial acetic acid, at 15 C., the oleic acid is just soluble, while the solid fat acids are insoluble. A greater proportion of the acetic acid precipitates oleic acid from its alcoholic solution ; a smaller proportion per- mits the solution of stearic and palmitic acids. A weighed portion of one or two grams of the fat acids under examination, in a flask provided with a tight stopper, is treated with the sol- vent mixture, in twenty-four hours' digestion at about 15 C., with occasional shaking. The mixture is then filtered, washed first with the solvent mixture, then with cold water, gathered into a weighed dish, melted, drained of water, dried in a desic- cator or at 100 C., and weighed as stearic acid. f. Quantitative. Free stearic acid, in absence of other acids, or a total of fat acids to be estimated as stearic acid, may be de- termined in quantity by acidimetry, using phenol-phthalein or litmus as an indicator, and taking the fat acid in alcoholic solu- tion. Each c.c. of normal solution of alkali represents 0.284 gram of stearic acid. Taking 2.84 gram of the material under estimation, each c.c. of decinormal solution of alkali equals 1 per cent, of free stearic acid. Free stearic acid, as obtained by precipitating an alkali stear- ate in aqueous solution with a diluted acid, washing with water, melting to separate water, and drying, may be weighed as C 18 H 36 O 3 . Also the residue of its ethereal solution may be melted, brought to a constant weight, and weighed in the same way. From Oleic acid stearic acid is separated as directed under 0, p. 242 ; in mixture with Palmitic acid stearic acid is estimated by methods given under Fat Acids, Quantitative Determinations of, (5) and (7), p. 250. g. Stearic acid is the u stearin " of the candle industry. 1 Ding. pol. Jour., 231, 64; Zeitsch. anal. Chem., 18, 622; Benedikt's "Analyse der Fette " (1886), p. 81; Am. Jour. Phar., 55, 356. 244 FATS AND OILS. For determinations of commercial value see under reference last given, especially methods (4) to (8). PALMITIC ACID. C 16 H 32 O 2 = 256 (monobasic). Margaric Acid. 1 Found as a normal glyceride in ordinary vegetable and animal fats. a. As free acid, crystallizable from alcoholic solution in fine needles, sometimes grouped in sheaves, or congealing from a melted mass in partly crystalline forms of pearly lustre. Melts at 62 C., at which temperature the liquid has the sp. gr. 0.852T. At about 350 C. it distils with partial decomposition. It leaves a permanent oil stain on paper. Under the reduced pressure of 100 millimeters it distils at 268.5 C. The glyceride, Palmitin, C 3 H 5 (C 16 H 31 O 2 ) 3 , is crystallizable, in pearly lustrous forms. It melts at temperatures from 50.5 to 66.5 C., according to its previous exposure fo heat. Strongly heated it carbonizes abundantly. ~b. Palmitic acid, as well as palmitin, is odorless and of a bland, oily taste. c. Palmitic acid is insoluble in water, and but sparingly and slowly soluble in cold alcohol, requiring 10.7 parts of absolute alcohol for solution, but hot alcohol dissolves it more freely, yielding crystals as the solution cools. It dissolves freely in ether. The alcoholic solution has an acid reaction. The normal glyceride, palmitin, is but slightly soluble in cold alcohol, moderately soluble in hot alcohol, and soluble in ether, chloroform, benzene, petroleum benzin, and in carbon disulphide. Alkali palmitates (soaps of palmitin) are soluble in water, with tendency to turbidity from slight decomposition, increased by dilution ; more permanently soluble in alcohol, scarcely at all soluble in ether. ^Ton-alkali metallic palmitates are insoluble in water or ether. Lead palmitate is insoluble in alcohol. Barium and calcium palmitates are slightly soluble in alcohol. d. In qualitative reactions palmitic acid does not sensibly differ from stearic acid. Its distinction from stearic acid re- quires quantitative work. e. Separations of palmitic acid are made with stearic acid, or, if this be absent, by the same methods given for stearic acid sep- aration (Stearic Acid, e). 1 This synonym is used by some French chemists. VARIOUS ACIDS. 245 y. Quantitative determinations of palmitic acid alone are done by the methods given for Stearic acid. When in mixture with stearic acid, methods of indirect determination are resorted to, as given under Fatty Acids, Quantitative Estimation of, (5) and (7). g. Palmitic acid enters into the stearic acid known in com- merce and in candle manufacture as " Stearin," and into " Oleo- margarin." See under these heads (Index). MYRISTIC ACID, C 14 H 28 O 2 . The fourteen-carbon member of the C n H 2n O 2 series of fat acids. Closely resembles Laurie acid. A solid, melting at 53.8 C., at which temperature the liquid has sp. gr. 0.8622. It is insoluble in water, sparingly soluble in cold alcohol and in ether. LAURIC ACID. The C-^IL^Og acid obtained from fats is a solid, fusible at 48.6 C., and of sp. gr. 0.883 at 20 C. It crys- tallizes from alcohol in needles. It does not vaporize, alone and under ordinary pressure, without being mostly decomposed, but distils with steam. In large quantities of boiling water sensible traces are dissolved. CAPRIC ACID, C 10 H 20 O 2 . The capric acid obtained from fats is solid at ordinary temperatures, forming small tabular crystals, melting at 31.3 to 31.4 C., boiling at 268-270 C., and of sp. gr. 0.93 at 37 C. It is soluble in about 1000 parts of water ; its calcium salt, very slightly soluble in water. CAPRYLIC ACID, C 8 H 16 O 2 . The caprylic acid obtained from fats congeals at 12 C. to a crystalline mass, melting at 16.5 C. Boils at 236-237 C. At 20 C. has sp. gr. 0.914. Of a sweet taste. Soluble in 400 parts of water. The calcium salt dissolves in 200 parts of water. CAPROIC ACID, C 6 II 12 O 2 . Isobutyl-acetic acid. Found as a glyceride in fats. Congeals at 18 C., boils at 199.7 C., is scarcely at all soluble in water. At 20 C., sp.gr. 0.925. Of a sweetish taste. The calcium salt dissolves in 37 parts of water. FATTY SERIES OF ACIDS, C n H 2n _ 2 O 2 . Oleic acid series. The following members of this and other immediately related series are found in fats : 246 FA TS AND OILS. Oleic Acid ....... C 18 H 34 2 = C 17 H 33 .CO 2 H. SERIES C n H 2n _ 2 O 3 : Ricinoleic Acid. . . C 18 H 34 O 3 . SERIES Onll^.^ : Linoleic Acid.... C 16 H 28 O 2 = C 15 H 27 .C0 2 H. OLEIC ACID, C 18 H 34 O 2 = 282. The members of the fatty series C n H 2n _ 2 O 2 contain two atoms of hydrogen less than cor- responding members of the fatty series C n H 2n O 2 , and by action of reducing agents the former are in general convertible into the latter. The normal glyceride of oleic acid, olein, C 3 H 5 (C 18 H 33 O 2 ) 3 =: 884, is found in greater or smaller proportion in most vegeta- ble and animal fats, and in non-drying oik. a. Pure oleic acid is a colorless oil, congealing at 4 C., melt- ing at 14 C., at which temperature the liquid has sp. gr. 0.898. Under ordinary pressure it does not distil alone undecomposed, but is carried over with superheated steam at about 250 0. The triglyceride, olein, is a liquid which congeals at low atmos- pheric temperatures, and in vacuum distils slowly without de- composition. b. Oleic acid is a bland, tasteless, when pure nearly or quite odorless liquid, indifferent in physiological action. c. Oleic acid is insoluble in water, soluble in alcohol not very dilute, and separated from the solid fat acids by its greater solubility in a mixture of acetic acid and alcohol. It is soluble in chloroform, benzol, petroleum benzin, and in the fixed oils. The triglyceride, olein, is somewhat soluble in absolute alcohol, in fact much more so than are stearin and palmitin, but is inso- luble in dilute alcohol. Pure oleic acid is neutral to litmus-pa- per, but it gives the acid reaction with phenol-phthalein, decolor- ing the alkaline mixture of this indicator at formation of normal alkali oleates. By exposure to air for a short time oleic acid suffers such changes as impart to it an acid reaction, and it soon becomes rancid and of a yellowish color. The alkali oleates are soluble in water, the solution becoming somewhat turbid by de- composition when diluted with water, though bearing dilution better than stearate or palmitate. The oleates of non-alkali me- tals are insoluble in water, but more or less freely soluble in al- cohol, and in some instances (including the lead salt) soluble in O LEI C ACID. 247 ether. The silver oleate is not soluble in ether. The alkali oleates are precipitated from their aqueous solutions by sodium chloride, and to some extent by excess of alkalies. Sodium oleate is soluble in 10 parts of water at 12 C., in 20.6 parts of alcohol of 0.821 applied at 13 C., or in 100 parts of boiling ether. From absolute alcohol it is crystallizable. Potassium oleate, in ordi- nary moist condition, is soft or gelatinous, and is much more so- luble in water or alcohol or ether than is the sodium salt. Ba- rium oleate is insoluble in water, and but very slightly soluble in alcohol. Magnesium-alkali oleate possesses a capacity of slight and transient foaming in aqueous solution, perhaps due to a tardy precipitation, and distinguishing it from calcium oleate in the soap test of hard waters. d. Oleic acid is characterized by its consistence as a liquid non-volatile fatty body, and by the action of oxidizing agents upon it. Nitric acid with metallic copper, fuming nitric acid, mercury nitrates, or other form of nitrous acid, in digestion with oleic acid, produces its isomer elaidic acid, as in digestion with olein it forms elaidin, glyceride of elaidic acid. Elaidic acid is a solid, and its formation is indicated first by a soft waxy, and finally by a resinous consistence. Elaidic acid dissolves in alco- hol, from which it crystallizes in tabular forms, melting at 45 C. Bromine acts readily, and iodine or chlorine quite easily, on oleic acid, producing dibrom-stearic acid, an addition product of oleic acid, C 17 H 33 Br 2 . CO 2 H, on the type of the C n II 2n O 2 series. To 7 parts of the oleic acid 4 parts of bromine are added, drop by drop, stirring after each addition. The product is yellowish, of an oily consistence. To form the di-iod-stearic acid, molecular proportions of the oleic acid and of the iodine are taken, each being dissolved in alcohol, when the iodine solution is gradually added, this being the reaction of Hiibl's estimation, giving the iodine number. e. Separations. In manufacture oleic acid is separated from the solid fat acids by filtration under pressure at low tempera- tures above the congealing point of the oleic acid. For methods of separation from the solid fat acids by sol- vents, etc., see Stearic acid, e. Directions for separation (pro- duction) from olein by saponification are essentially those given under Hehner's method. f. Quantitative. Oleic acid is estimated volumetrically by standard solution of potassa or soda, using phenol-phthalein as an indicator. Each c.c. of normal solution of alkali represents 248 FATS AND OILS. 0.282 gram of oleic acid. Taking 2.82 grams of material, each c.c. of decinormal solution of alkali counts 1 per cent, of oleic acid. Taking 14.1 gram of material, c.c. of normal solution of alkali X ! 2 = per cent, oleic acid. Gravimetric estimation of free oleic acid is effected by adding to the free acid, in a layer over an aqueous liquid, a weighed portion of recently fused beeswax or paraffin, heating to melt the solid, and when cold detaching the oily mass, drying in a tared capsule, and weighing, when the weight of the wax is subtracted. Also free oleic acid, dissolved in ether, may be freed from the latter by evaporation in a tared beaker or flask, avoiding oxidation by exposure to the air, and the weight of the oleic acid may be obtained. g. The U. S. Ph. gives the following specifications for oJeic acid : " At 14 C. (57 F.) it becomes semi-solid, and remains eo until cooled to 4 C. (39 F.), at which temperature it becomes a whitish mass of crystals. At a gentle heat the acid is completely saponified by carbonate of potassium. If the resulting soap be dissolved in water and exactly neutralized with acetic acid, the liquid will form a white precipitate with test solution of acetate of lead. This precipitate, after being twice washed with boiling water [drained and dried], should be almost entirely soluble in ether (absence of more than traces of palmitic and stearic acids). Equal volumes of the acid and alcohol, heated to 25 C. (77 F.), should give a clear solution, without separating oily drops upon the surface (fixed oils)." The specifications of the Br. Ph. are as follows : " A straw-colored liquid, nearly odorless and tasteless, and with not more than a very faint acid reaction. Unduly ex- posed to air it becomes brown and decidedly acid. Specific gravity 0.860 to 0.899. At 40 to 41 F. (4.5 to 5.0 C.) it be- comes semi-solid, melting again at 56 to 60 F. (13.3 to 15.5 C.) It should be completely saponified when warmed with carbonate of potassium, and an aqueous solution of this salt neutralized by acetic acid and treated with acetate of lead should yield a preci- pitate which, after washing with boiling water, is almost entirely soluble in ether." RICINOLEIC ACID, C 18 H 34 O3=298. The fat acid constituting, in its normal glyceride, the principal part of castor oil. In com- position an oxy-oleic acid of the proportions C n H 2n _ 2 O 3 . a. A thick oil, of sp. gr. 0.940 at 15 C., congealing at 6 to 10 C., and does not vaporize undecomposed. The lead salt melts at 100 C. b. The glyceride, as obtained in castor oil, is odorless, with a LINOLEIC ACID.HYPOGAIC ACID. 249 mild taste and slightly acrid after-taste, and exerting a cathartic effect. c. Ricinoleic acid is insoluble in water ; soluble in all pro- portions in alcohol and in ether. The lead salt is soluble in ether. Castor oil is soluble, at 15 C., in 2 parts of 90$ alcohol or 4 parts of 84$ alcohol. It is but slightly soluble in petroleum benzin, paraffin oil, or kerosene, though it takes up about its own volume of petroleum benzin. d. Ricinoleic acid is but very slightly oxidized by exposure to air. Treated with bromine it takes two atoms of bromine in the molecule of the acid, forming C 18 H 34 Br 2 O 3 , corresponding to the reaction of oleic acid. In the elaidin reaction, by action of nitric acid, ricinelaidic acid is formed, isomeric with ricinoleic acid, and fusible at 50 C. LINOLEIC ACID, Ci 6 H 28 O 2 252. The only well-known mem- ber of the series of fat acids O n H 2n _ 4 O 2 . In the normal glyce- ride forms the principal portion of linseed oil, representative of the drying oils. In oxidation, or " drying," it forms addition products, such as C 16 H 28 Br 4 O 2 , corresponding in composition to C n H 2n O 2 . Therefore .in reaction with oxidizing agents it has twice the saturating power per molecule possessed by oleic acid. a. Linoleic acid is a permanent liquid of a pale yellow color and sp. gr. 0.9206 at 14 C. The glyceride, as obtained in linseed oil, congeals at 16 C. (GussEBow), 27 C. (AECHBUTT and AL- LEN), and melts at 16 to 20 C. (GLASSNEE). b. Linseed oil has a characteristic odor and taste. G. Insoluble in water, readily soluble in alcohol and in ether. Of a feeble acid reaction, and capable of neutralizing alkalies to phenol-phthalein and other indicators. The barium and calcium salts dissolve in hot alcohol. Ether dissolves the linoleates of lead, zinc, copper, and calcium. d. Linoleic acid is easily oxidized by exposure to the air. In thin layers in a few days it forms a solid, resin-like body known as Oxylinoleic acid, and afterward takes on the character of a neutral body, insoluble in ether, and sometimes termed Linoxyn. HYPOGAIC ACID, C 16 H 30 O 2 . A white solid, crystallizing in needles, melting at 33 C., not readily vaporizing in ordinary conditions without decomposition. By exposure to air soon be- 250 FATS AND OILS. corned rancid, acquiring a brown color, and giving origin to vola- tile acids. In the elaidin test it is changed to its isomer gaidic acid, fusible at 39 C. PHYSETOLEIC ACID, isomeric with Hypogaic acid, C 16 H 30 O 2 , melts at 30 C., and is not affected by the elaidin test. FAT ACIDS, QUANTITATIVE DETERMINATIONS OF. Besides the methods of volumetric and gravimetric estimation of separate fat acids, or of total fat acids in terms of stearic acid, by equivalence of saturation, certain special determinations have been made, upon stated authorities, as indices of composition, related to as- certained limits, representing values for given uses. (1) The number of parts of insoluble fat acids obtainable from 100 parts of clear neutral fat (HEHNER'S number). (2) The number of c.c. of decinormal alkali solution saturated by the volatile fat acids distilled from 2. 5 grams of the fat (REICHERT'S number). (3) The number of milligrams (thousandths) of potassium hydrate saturated by saponifying 1 gram (one part) of the fat (KOTTSTORFER'S number). The sapoimication number. The methods above named have been devised to distinguish butter from its substitutes. (4) The percentage of iodine which the oleins of the fat will take into combination by a defined procedure (HUBL'S iodine numberX (5) The molecular weight, as obtained by acidimetry. (The quantity saturated by 1000 c.c. normal solution of alkali.) For mixtures of palmitic and stearic acids. (6) The specific gravity, as a limited indication. (7) The melting and congealing points. (8) Calculation of constituent Fat Acids and Neutral Fats. (1) Estimation of the insoluble fat acids : HEHNER'S Method. 1 To obtain the butter fat from butter melt a portion on the water-bath, leave the liquid to settle while melted, decant the ], 7, 131. JONES, 1877: Analyst, 2, 20. FLEISCHMANN and VIETH. 1878: Zeitsch. anal. Chem., 17, 287. Manipulation at the Depart, of Agriculture, Washington, REPORT DEPT. AGR., 1884, Prof. WILEY, Chemist, p. 60. Further, see citations under Butter Fat. DE TERMINA TION OF FAT A CIDS. 2 5 1 the water-bath. Keep in a light beaker, and take out for an analysis from 3 to 4 grams of the clear fat into an evaporating- dish of about 5 inches (13 centimeters) diameter, using a glass rod to be left in the evaporating-dish, and weighing tli^e beaker before and after the removal to obtain the exact weight of fat taken. Add 50 c.c. of alcohol of about 85$, and 1 to 2 grams of pure (alcoholic) potassium hydrate, and warm and stir the mix- ture until a clear solution is obtained. After five minutes' fur- ther warm digestion add a few drops of distilled water, and if a turbidity is caused continue the digestion until the addition of water produces no turbidity. If this satisfactory saponification is not attained the failure is probably due to a too great dilution of the potash with alcohol, and the operation is to be commenced anew. If the alcohol be too strong, saponification is prevented. The clear saponified solution is now evaporated over the water- bath to a syrupy consistence, and the residue dissolved in 100 to 150 c.c. of water. To the clear liquid add diluted sulphuric or hydrochloric acid to a strongly acid reaction. The creamy sepa- rate of the insoluble fat acids rises for the most part to the sur- face. Heat of a bath of water below boiling is now applied to melt the precipitate, and continued for half an hour and until the layer of fat acids above is perfectly clear and the aqueous liquid below is nearly clear. Meantime a filter of 4 to 5 inches (10 to 13 centimeters) diameter, of the closest filter-paper (Swe- dish), is dried in the water-box. The filter should be close enough to transmit hot water only by drops. A small beaker is weighed, also a filter weighing-tube and this tube with the filter, to give the weight of the latter. The weighed filter is placed in a funnel wetted and half -filled with water. The watery liquid and melted fat are then poured from- the dish upon the filter, which is not to be at any time more than two- thirds filled ; the dish and rod are rinsed with boiling water, and washing with boiling water is to be continued until the washings cease to redden liimus-paper, about f liter (700 to 1000 c.c.) of filtrate being usually obtained. 1 (The rins- ing of the dish seldom leaves behind more than a milligram of fat, but this is saved by taking it up with a little ether and the solution added to the fat acids in the beaker afterward.) The 1 FLETSCHMANN and VIETH (1878) advise care to avoid imperfect solution of lauric acid (abounding in cocoanut oil), washing until 5 c.c. of the filtrate ceases to change the color of one drop of litmus tincture added thereto. E. WALLER and his associates (1886: Report N. Y. State Dairy Commissioner) wash with six or seven instalments of hot water (about 100 c.c. each), rinsing off between each with about 25 c.c. of cold water. 252 FATS AND OILS. drained funnel is set well down in a beaker of cold water, and when the fat acids have hardened the filter is detached, drained, and placed in the weighed beaker. ' This is heated on the water- bath to a (nearly) constant weight. Weigh after about two hours' drying, and after a half-hour's further drying weigh again. If any drops of water collect below the fat add a drop or two of alcohol. In this drying there may be slight increase by oxida- tion of oleic acid, and slight decrease by vaporization of fat acids. If the filter have been close enough no fat globules will have passed, and none will be revealed by microscopic examination of the filtrate. The weight of the beaker and contents, minus the weights or tares of the beaker and the filter, leaves the weight of the fat acids, which is to be divided by the weight of purified fat taken, to obtain the proportion ( X 100 = $) of insoluble fat acids. If 87.5 be accepted as the full per cent, of insoluble fat acids in butter, and 95.5 as the per cent, of insoluble fat acids in " meat fats," then 95 5 87.5 = 8, and 8 : found percentage minus 87.5 :: 100 : a? = per cent, of " meat fats" present in the clear fat examined. For the calculation of percentage in entire but- ter see under Butter, Interpretation of Results. DALICAN modifies Hehner's process by taking 10 grams of the clear butter fat in a flask of 250 to 300 c.c. capacity, and adding 80 c.c. of 85$ alcohol, and 6 grams of sodium hydrate dissolved in 6 to 8 c.c. water, when by 30 to 40 minutes of warming and stirring the saponification is ended. The alcohol is evaporated off, 150 c.c. of water added, and 25 c.c. of hydrochloric acid diluted with four parts of water are added in small portions at a time, rotating the flask after each addition. The mixture is now heated over the water-bath for 25 to 30 minutes, until the fat layer separates with perfect clearness and white points are no longer seen. The flask is set aside for 30 minutes, and then cooled with water. After two hours the fat layer is broken with a glass rod, the water poured on a wetted filter, about 250 c.c. of boiling water added in two portions to the flask, shaking after adding the first portion. The flask is then set aside 40 minutes, cooled by immersion in water, and the water decanted on the filter as before. This washing by decantation, as above, is repeated until the decanted liquid ceases to redden litmus- paper on 20 minutes' contact, 8 or 10 washings being necessary. 1 "The insoluble acids are brought into a tared dish, any in the filter or flask being dissolved in ether, dried at 100 C. with stirring with absolute alco- hol to remove water, and weighed." H. W. WILEY, Chemist Dept. Agricul- ture, Washington, Report of 1884. DE TERMINA TION OF FAT A CTDS. 2 5 3 The insoluble fat acids are finally gathered in a tared porcelain capsule or evaporating-dish, the flask being washed with hot water, and all washings passed through the filter. The filter must be kept wet, and the slight portion of fat acids upon it can easily be detached. The drying is done at 100 to 110 C., at first for an hour, and a second weight is taken in 15 or 20 minutes. For results with vegetable and animal fats see tables following ; also Butter Fat. (2) REICHERT'S method 1 embraces the estimation of the vola- tile fat acids, separated by distillation. " Reichert's number" is the number of c c. of decinormal solution of alkali taken to neu- tralize the distilled fat acids from 2.5 grams of fat. Sometimes, however, results are specified for 5 grams or for 10 grams of the fat. Of the clear filtered fat 2.5 grams are taken in an Erlen- meyer's flask of about 150 c.c. capacity, with 1 gram potassium "hydrate and 20 c.c. of 80$ alcohol, and the whole digested on the water-bath, with shaking by circular motion until saponification is complete and the alcohol is all removed. Now 50 c.c. of water are added, then 20 c.c. of diluted (1 to 10) sulphuric acid, and the mixture distilled. To avoid bumping a slight stream of air may be introduced. The distillate is received in a 50 c.c. flask, into which is set a funnel carrying a wetted filter, receiving the distillate, so that any insoluble fat acid otherwise possible in the distillate may be rejected. The first 10 or 20 c.c. of distillate are returned to the flask ; then 50 c.c. are distilled. The volume of the distilled liquid should always bear the same proportion to that of the distillate. This distillate is charged with a few drops of phenol-phthalein solution, and titrated with the decinormal solution of alkali, until the color of the alkali reaction becomes constant. The required number of c.c. (2.5 grams of fat having been taken) is Reichert's number. MEISSL'S modification of Reichert's process 2 undertakes a more complete distillation of the volatile fat acids, and the use of weaker alcohol in saponifl cation to avoid etherizing the acids, as follows : 5 grams of the clear filtered fat are treated in a flask of 200 c.c. capacity with 2 grams of solid potash and 50 c c. of alcohol (free from acidity or aldehyde), over the water-bath, 1 E. REICHERT, 1879: Zeitsch. anal. Chem., 18, 68; Jour. Chem. Soc., 36, 406. ALLEN, 1885: Analyst, 10, 103. R. W. MOORE, 1885: Jour. Am. Chem. Soc., 7, 188; Analyst, 10, 224; Am. Chem. Jour., 6, 417: Chem. News, 50, 268; Jour. Chem. Soc., 48, 300, 1014. E. REICHAEDT, 1884: Zeitsch. anal. Chem., 23, 565; Jour. Chem. Soc., 46, 1219. 2 E. MEISSL, 1880: Bied. Cent., 1880, 471; Jour. Chem. Soc , 38, 828. 254 "FATS AND OILS. with stirring, until saponified perfectly. The alcohol is evapo- rated, and the thick soap is dissolved in 100 c.c. of water, pre- cipitated with 40 c.c. of diluted (1 to 10) sulphuric acid, and, after the addition of a few pieces of pumice-stone, distilled with use of a Liebig's condenser. Of distillate 110 c.c. are received in a flask marked at this capacity, this quantity being obtained in about an hour. The distillate is filtered into a flask marked at 100 c.c., and this volume of the filtrate is titrated, after addition of phe- nol-phthalein or litmus, with decinormal solution of alkali. The number of c.c. required is increased by its one-tenth, and for Reichert's number (on 2.5 of fat) the result of this operation is divided by two. To exclude all interferences a control analysis without fat may be conducted parallel with the assay. The Reichert's numbers of fats are given under Butter Fat. (3) Kottstorfer 1 s method l Determination of the number of milligrams of potassium hydroxide necessary to saponify 1 gram of the fat the " Yerseif ungszahl," or saponification number. The operation requires (1) solution of hydrochloric acid, and (2) alcoholic solution of potassa, both of about half-normal strength. Also (3) a decinormal solution of alkali, exactly standardized. The potassa solution is made of caustic potassa purified by alco- hol, dissolved in the least sufficient proportion of water and di- luted to standard with alcohol free from fusel oil. It may be prepared by filtration through 'animal charcoal. If the alcohol be pure a solution of the designated strength will not become darker than yellowish. Of the purified fat 1 to 2 grams are digested in a covered beaker or flask of about 70 c.c. capacity, with just 25 c.c. of the alcoholic potassa solution, on the water-bath, at near boiling of the liquid, stirring with a glass rod, to perfect saponification. It is believed to be necessary to take precautions against the escape of ethyl butyrate. One c.c. of phenol-phthalein solution is added, and the liquid titrated with the standard hydrochloric acid to the neutral point. Another 25 c.c. of the potash solution alone is titrated with the hydrochloric acid solution, and the latter titrated with the deci- normal alkali. The number of c.c. of the standard alkali taken for the 1 gram of fat, minus the hydrochloric acid in the titra- tion, converted, according to the comparisons made, into milli- 1 J. KOTTSTORFER, 1879: Zeitsch. anal. Chem., 18, 199, 431; Jour. Chem. Soc., 36, 983, 1069; Analyst, 4, 106. MOORE, 1885: Jour. Amer. Chem. Soc., 7, 188; Analyst, 10, 224; Chem. News, 50, 268; Am. Chem. Jour., 6, 417; Jour. Chem. tioc., 48, 300, 1014. DETERMINATION OF FAT ACIDS. 255 grams of potassium hydroxide, gives the saponification number sought. The details are carried out upon fats of butter and its substitutes by Prof. WILEY (1884) as follows : The dried and filtered butter-fat is weighed in a small beaker with a 2 c.c. pipette. Five stout half-pint beer-bottles of clear glass, with rubber stoppers secured by a spring, are provided. Three portions, of 2 c.c. each, of the fat melted at about 35 C. are introduced severally into three of the bottles, weighing the beaker and pipette after each addition, and noting the exact weight of fat taken in each of the three bottles. Of the alco- holic potash solution just 25 c.c. is now run into each of the five bottles. The bottles are stoppered and placed on the same steam or- water-bath, and shaken every five minutes until the fat is sa- ponified. Then the bottles are cooled, opened, and 1 c.c. phenol- phthalein solution added to each. Each of the five portions is now titrated with the half -normal hydrochloric acid to the neu- tral point. The two blanks give an average for the strength of the alcoholic potash, and the three portions of fats give an ave- rage for the amount of potash neutralized in saponification. That is, the mean number of c.c. of hydrochloric acid for one of the two blank portions, minus the mean of c.c. of the same acid calculated for 1 gram of fat in one of the three fat-portions, equals the no. c.c. of the hydrochloric acid neutralized by the total fat acids in 1 gram of the fat. This last no. of c.c. of hydrochloric acid is to be titrated with the exactly standardized decinormal al- kali, and the required no. of c.c. of the latter is multiplied by 5.6 to obtain the milligrams KOH for the fat acids of 1 gram of fat. Kottstorfer's Number, the milligrams of KOH to saponify 1 gram of fat, is to be distinguished from the " Saturation-Equi- valent " of fats. The latter term is defined as the number of milligrams of fat saponifiable by 1 c.c. of normal alkali solution. For the triglycerides it is the third of their molecular weight ; or it is the hydrogen-equivalent number of the fat. 56000 -T- Kottstorfer's number = " saturation-equivalent " ; and 56000 -r- " saturation -equivalent " = Kottstorfer's number. PEKKINS 1 combines the methods of Hehner, Reichert, and Kottstorfer, as follows : Of the clear fat 1 to 2 grams is saponified ; an excess of a cold-saturated solution of oxalic acid is added, and the fat acids separated in the cold, and then washed on the filter with hot water. The filtrate is made up to 200 c.c., and distilled 1 F. P. PERKINS, 1878: Analyst, 3, 241; Zeitsch. anal. Chem., 19, 237. 2 5 6 FATS AND OILS. to give 100 c.c. of distillate (according to Reichert), this being titrated with alkali and the result stated in milligrams of potas- sium hydroxide to saturate the volatile acids from 1 gram of pu- rified fat. The insoluble fat acids, as washed, are dissolved in 100 c.c. of hot alcohol, and this solution, or an aliquot part of it, titrated with decinormal alkali, calculating the result into milli- grams of potassium hydroxide for the insoluble fat acids of 1 gram of fat. The former number plus the latter number gives the milligrams of potassium hydroxide to saturate all the fat acids of the 1 gram of purified fat. Percentages of Insoluble Fat Acids. Hehner's Numbers. Olein. Palmitiii Stearin Eutyrin Oleomargarin. . Cotton- seed oil Cotton stearin Lard.. Olive oil Peanut oil .... Palm oil Sesame oil. . . . Theobroma oil Seal oil Rape oil Cocoanut oil . . Butter fat, lowest . . highest.. " common mum. maxi- 95.75 95.28 95.73 87.41 95.56 95.75 94.29 95.5 96.15 95.43 95.09 95.00 95.6 95.48 94.59 90.68 95.10 86.43 80.78 86.6 88.5 87.5 Theoretical quantity. HEHNER'S determinations. (BENSEMANN). (E. WALLER, 1886). (MUTER). (WEST-KNIGHTS). u (E. WALLER). a (HEHNER). (E. WALLER). (BENSEMANN). (E. WALLER). (BENSEMANN). (MOORE). (E. WALLER). (HEHNER and ANGELL). Butter fat, lowest of nine . . . " highest " ... From 26 genuine I lowest. American butters, j" highest From 25 butters, ) lowest. Pennsylvania . . . . j highest 88 89 jjg i WILEY, Washington, 1884. 86.40 ) E. WALLER, 90.24 j New York, 1886. 86.7 87.7 C. B. COCHRAN, 1886. DETERMINATION OF FAT ACIDS. 257 Saponification Coefficients : Kottstorfer* s Numbers (p. 254). (Milligrams of KOH neutralized in saponifying 1 gram of Fat.) Stearin 188.8 By calculation. Olein 190.0 " Palmitin 208.0 " Butyrin 557.3 " Beef Tallow 196.5 KOTTSTORFER. k < " commercial 196.8 " Mutton Tallow 197.0 " Lard 195.7 " Olive oil 191.8 " Eape " 178.7 " (mean... 227.0 Butter Fat. ...'... -I lowest. . 221.5 " ( highest. 233.0 Fat of Rancid Butter about 1.5 lower than when fresh. " Cocoanut oil 250.3 (MooEE, 1884). " " washed 246.2 " " " 49.3$, Oleomar- garin 50.70 220.0 " Cocoanut oil 70.2$, Oleomar- garin 29.8$ 234.9 Almond oil, sweet 194.7-196.1 (YALENTA, 1883). Apricot oil 192.9 " Bitter Almond, fixed oil . . 194. 5 n -i j 181 - " Castor Ol1 j 176-178 (ALLEN, 1884). Cotton-seed oil 195 (YALENTA). Lard oil 191-196 (ALLEN). T . , ., 189-195 " Lmseed011 195.2 (MOORE). 191.7 (YALENTA). Olive oil 191-196 (ALLEN). 185.2 (MOORE). p n j 191.3 (YALENTA). Peanut011 I 196.6 (MooBE). Sesame oil 190.0 (YALENTA). Sperm oil 130.0-134.4 (ALLEN). Theobroma oil 199.8 (MOORE). Train oil 190-191 (ALLEN). Messrs. WALLER and MARTIN (1886 : Report of the Dairy 258 FATS AND OILS. Commissioner of the State of New York) obtained, from 25 genu- ine American butters, Kottstorfer's numbers from 220.6 to 230.1 (extremes) ; a rancid butter, 223.0 ; the same deodorized, 219.45 ; and from the insoluble fat acids of a butter, 214.25. From oleo- margarin 188.65; another, 191.6. From mutton suet, 203.25 ; beef suet, 199.2; lard, 195.85. From cottonseed oil, 162.0 to 193.05 ; average of five, 183.47. (4) Determination of Fat Acids by their capacity of combi- nation with iodine. The fat acids, whether free or in their glycerides, form combinations with iodine, bromine, or chlorine. One molecule of oleic acid or ricinoleic acid takes two atoms of iodine ; one molecule of linoleic acid, four atoms of iodine ; ad- dition products being formed. The directions of HUBL 1 for finding the percentage of iodine taken into combination (the iodine number) are as follows, com- mencing with preparation of the needful reagents : (I) Iodine solution. Of iodine 25 grams are dissolved in 500 c.c. of alcohol (free from fusel oil); of mercuric chloride 30 grams are dissolved in 500 c.c. of the alcohol, and this solution filtered if necessary ; when the two solutions are united, and, after 6 to 12 hours' standing, titrated with the standardized thiosulphate solution, and the standard noted. (2) Thiosulphate solution. A solution of about 24 grams of sodium thiosulphate in the liter is made, and its iodine value accurately determined with a weighed quan- tity of freshly sublimed iodine. About 0.2 gram of resublimed iodine is placed in a small glass tube closed at one end and pro- vided with a similar tube enough larger to serve as a cover, both tubes being previously dried and weighed. The iodine is heated in the inner tube, on a sand-bath, until it melts, then covered with the outer tube, cooled in a desiccator, and weighed. The cover is now removed and both tubes are placed in a stoppered flask con- taining 1 gram of potassium iodide (neutral and free from iodine) dissolved in 10 c.c. of water. When the iodine has dissolved, the solution of thiosulphate of sodium is added from a burette until the iodine color is reduced to a faint yellow, a little starch solu- tion is added, and the titration completed to the extinction of the blue color. The iodine value of the thiosulphate solution is now written. (3) Chloroform. The purity of chloroform is as- sured for this assay by digesting 10 c.c. of it with 10 c.c. of the iodine solution at ordinary temperature for two or three hours' and titrating to the extinction of the iodine with the thiosul- 1 1884: Ding. pol. Jour., 253, 281; Jour. Chem. Soc., 46, 1435; Am. Ghem. Jour., 6, 285. DETERMINA TION OF FAT ACIDS. 259 phate solution, the stated quantity of which should be consumed. (4) Potassium iodide solution. One part of pure iodide of potas- sium in 10 parts of water. It should be neutral in reaction, and should not contain any free iodine. For the assay, 0.2 to 0.3 gram of a drying oil, or 0.3 to 0.4 gram of a non-drying oil, or 0.8 to 1.0 gram of a solid fat, is taken in a close-stoppered fiask of about 200 c.c., and 10 c.c. of the chloroform are added for solution. Of the iodine solution 20 c.c. are added in exact measure, and, if the mixture does not become clear after shaking, a little more chloroform is added. The Quantity of iodine should be sufficient to leave a dark brown color after one and a half or two hours' standing, the time to be taken for the reaction. In titrating the remaining excess of the iodine, 10 to 15 c.c. of the potassium iodide solution and, after shaking, 150 c.c. of water are added, when the thiosulphate solu- tion is added, with shaking, until the color of both the aqueous layer and the chloroform layer is reduced to a pale yellow, when starch solution is introduced and the extinction of the iodin \ completed. For close results the iodine and thiosulphate solu- tions should be standardized just before or after the assay. The number of parts of iodine taken by 100 parts of the fat is known as its iodine number. Using a sufficient excess of iodine in the reaction, quite constant results are promised. Iodine Numbers. rn -A n TT r\ f 90 07 By calculation. (Jleic acid, (Jiolio^Uo OA , n ~ - - n j ( 89.8 to 90.5 By experiment. Bicinoleic acid, C 18 H 34 O 3 . . . 85.24 By calculation. Linoleic acid, C 16 H 28 O 2 201.59 " " Linseed oil 158 (HUBL). 155.2 (MooRE). 1 Hempseed oil 143 " Walnut oil 143 " " " Cotton- seed oil 106 it 108.7 " Sesame oil 106 u 102.7 " Rape and Rubsen oils .... Olive oil 100 82.8 u c< 103.6 " 83.0 " Olive-seed oil . 81.8 u Castor oil 84.4 u Almond oil (sweet) 98.4 u 98.1 " Mustard oil (fixed) 96.0 R. W. MOORE, 1885: Am. Chem. Jour., 6, 416. 2 6o FATS AND OILS. Bone oil 68.0 (HUBL). Cod-liver oil 123 to 141 (KREMEL). Lard . . 59.0 (HUBL). 61.9 (MOORE). Oleomargarin 55.3 50.0 Palm oil 55.5 " 50.3 Tallow... 40.0 " Wool fat. . , 36.0 Cacao butter (theobroma oil) 34.0 Mace oil (nutmeg butter). . 31.0 " Butter fat 31.0 " 32.8 to 38.0 " " " very old 19.5 " Cocoanut oil 8.9 " 8.9 " Japan wax 4.2 " Fat acids of bone oil 57.4 (MOBAWSKI and DEMSKI). Fat acids of tallow of beef. 25.9 to 32.8 " " Fat acids of cocoanut oil. . 8.4 to 8.8 " " Fat acids of linseed oil .... 155.2 to 155.9 " Fat acids of olive oil 86.1 " Fat acids of cotton seed oil. 110.9 to 111.4 Fat acids of castor oil 86.6 to 88.3 " " Mineral oils (petroleum, shale) Seldom above 14 YALENTA (1884). Eosin oils. 43 to 48 " Messrs. WALLER and MARTIN (1886 : Eeport of N. Y. State Dairy Commissioner) found Htibl's numbers as follows : Ayrshire butter 34.7 Jersey " sweet cream 36.7 " ' " sour cream 30.5 Native " 30.5 Devon " sour cream 37.0 Eancid 40.5 Oleomargarin 50.9 to 54.9 Cotton-seed oil 108.4 Linseed oil (average of 2). . . 165.4 Cocoanut oil (average of 2) 7.8 Commercial Stearin.. 1.7 DETERMINATION OF FAT ACIDS. 261 (5) Estimation of Stearic and Palmitic Acids separately in mixtures of the two, by the mean molecular weight of the mix- ture. The molecular weights of stearic and oleic acids, 284 and 282, do not differ from each other enough to give any value to this method applied to mixtures of these two acids. It is appli- cable to tallows from which the olein has been removed by pres- sure. About 50 grams are treated for the separation of the fat acids, by digesting with 40 c.c. potassium hydrate solution of sp. -gr. 1.4, and 40 c.c. of alcohol. After boiling to full saponification, one liter of water is added and the liquid boil- ed about three-fourths of an hour to remove the alcohol, diluted sulphuric acid is added to complete precipitation, the precipitate is well washed with water, melted until clear of water, drained and dried. An accurately weighed portion of about 5 grams of the clean fat acids is dissolved in alcohol, and titrated, by add- ing normal or other standard solution of alkali in some excess, using phenol- phthalein as an indicator, and bringing back the neu- tral point by a corresponding solution of acid, when the number (n) of c.c. of normal solution of alkali for saturation of 1 gram of the fat acids is found. Then 1000 -r- n = mean molecular weight. Now let x be the desired per cent, of the one fat acid, and b its molecular weight; y the desired per cent, of the other acid, and c its molecular weight while a is the mean molecu- Ct -r /* lar weight as found from titration. Then x = 100 = , and c y = 100 x. With stearic and palmitic acids x = 100 . 28 (6) Determination of Specific Gravity of the Fats. The determination of the liquid fats is made at customary standard temperatures, as of other liquids, by weight in a specific-gravity bottle, or a Sprengel's tube, or by a hydrometer. The specific gravity of the waxes and very hard fats is usually taken in the solid state, at customary temperatures, and so stated. In case of many fats, however, the density of the solid state is measurably dependent upon the conditions of solidification. And the speci- fic gravity of the softer solid fats is more often taken in the liquid state, at some stated temperature well above the melting point by Stoddart at 100 C., by Muter at 37.8 C. (100 F.) and usually taking water at the same temperature as the stand- ard. In adjusting the temperature of a specific-gravity bottle or a Sprengel's tube, immersion in water is employed. The water is warmed in a beaker, or other convenient vessel, in which the 262 FATS AND OILS. gravity bottle or tube is securely suspended, the filling up of the exact volume of the liquid fat being adjusted at the noted tem- perature. Then the operation is repeated with water instead of fat, to obtain the divisor representing the unit of density. The Westphal balance is conveniently employed, the counter- poise being suspended and weighed in the oil, contained in a vessel surrounded by water in a larger vessel to which heat is applied. To take the specific gravity of a melted fat by the hydro- meter, a small hydrometer is most convenient, and the oil is contained in a corresponding small cylinder which can easily be immersed in a hot water-bath. A constant water-bath of constant temperature is convenient for habitual use in this operation. Methods by dropping into liquids of known and adjusted density have been proposed by chemists. HAGER (1880) drops melted fat into alcohol, keeping the drops separate, then transfers them to mixtures of alcohol, water, and glycerine, until an equi- librium is found. A list of densities of fats and resins, so deter- mined, is published. 1 A similar method was proposed for butter fat, using methylated spirit, by CASSAMAJOR (1881), and further described under Butter. The counterpart principle, of employ- ing specific-gravity beads of graded density, has been developed by Mr. WIGNER (1876 3 ), especially for melted fats. BLYTH recommends taking the specific gravity of butter fat, clarified and filtered, as a solid at 15 C., by weight in suspension in water, with a weighted tube, on the general plan of solids lighter than water. 3 The ratio of expansion of butter fat, lard, etc., on increase of temperature, has been reported on by WIGNER (1879 4 ). SPECIFIC GRAVITIES or FAT OILS classified by BENEDIKT, tak- ing figures of ALLEN (1884), and others, at 15 C. : A. Sp. gr. under 0.883 . . 1. Liquid waxes from ma- rine animals : Sperm oil 0.875-0.883 , 2. Oils of unknown com- position : Shark oil 0.865-0.867 African fish-oil 0.867 l Zeitsch. anal. Chem., 19, 239; Jour. Chem. Soc., 38, 70; Phar. Jour. Trans., [3], 10, 287. 2 Analyst, I, 145. 3 1880: Analyst, 5, 76. 4 Analyst, 4, 183. DETERMINATION OF FAT ACIDS. 263 B. 0.883 to 0.912 Oil from cranial cavities : Liquid waxes with glycerides 0.908 C. 0.912 to 0.920 1. Non-drying oils : Almond oil 0.917-0.920 Peanut oil. 0.916-0.920 Olive oil 0.914-0.917 Mustard oil 0.914-0.920 2. Oils of marine animals. none. 3. Oils of land animals : Lard oil 0.915 Tallow oil 0.916 Neat-foot oil 0.914-0.916 Bone oil 0.914-0.916 D. 0.920 to 0.937 1. Vegetable oils : (a) Feebly drying, sp. gr. less than 0.930: Cotton-seed oil. . . 0.922-0.930 i Sesame oil 0.923-0.924 Sunflower oil 0.924-0.926 (J) Strongly drying oils : Hempseed oil 0.925-0.931 Linseed oil 0.930-0.935 Poppy oil 0.924-0.937 Walnut oil 0.925-0.926 2. Oils of marine animals : Cod-liver oil 0.923-0.930 3. Oils of land animals . . none. E. Sp. gr. above 0.937. . 1. Vegetable oils. Of ca- thartic effect: Croton oil 0.942-0.943 Castor oil 0.960 (Boiled linseed oil.) 2. Oils of land animals . . none. SPECIFIC GRAVITIES OF FAT OILS at 15 C. (MUNICIPAL LA- BORATORY OF PARIS, 1884) : Almond oil 0.917 Olive oil 0.9163 Peanut oil. , 0.917 " "common.. . 0.9163 264 FATS AND OILS. Colza oil 0.9154 Cotton-seed oil (white) . 0.9254 " " (brown). 0.930 Beechnut oil 0.922 Linseed oil. 0.9325 Cameline oil 0.9252 Walnut oil 0.926 Poppy oil 925 Tallow oil. . Sesame oil 0.9226 Norwegian whale oil. . . 0.9257 South Sea " " . American " " . Cod- liver oil (pale) . . " " " (brown) ISTeat-foot oil 0.9142 Sheep-foot oil 0.9187 . 0.9029. 0.927 0.925 0.928 0.9254 To compare the specific gravity of one oil with that of an- other oil (DONNY, 1864): Color the one oil with alkanet or other tinctorial matter, and, while both oils are at same tempera- ture, let fall a few drops of the colored oil into a portion of the other in a test-tube. SPECIFIC GRAVITIES OF SOLID FATS at 15 C. (DIETERICH, 1882) : Wax, white 0.973 Common Resin, " yellow 0.963-0.964 American 1.108 " Japan 0. 975 Common Resin, Ceresin, white . . . 0.918 French 1.104-1.105 " half white. 0.920 Theobroma oil . . . 0.980-0.981 " yellow . . . 0.922 Paraffin, medium. 0.913-0.914 Ozokerite, crude . 0.952 Tallow, beef 0.952-0.953 Spermaceti. . 0.960 " sheep 0.961 "Stearin" 0.971-0.972. SPECIFIC GRAVITIES OF SOLID FATS at 15 C. HAGER (1879) : Butter Fat, clarified 0.938-0.940 " several months old 0.936-0.937 Artificial butter 0.925-0.930 Lard, fresh 0.931-0.932 Tallow, beef 0.925-0.929 " sheep 0.937-0.940 Cocoanut oil, fresh 0.950-0.952 " very old 0.945-0.946 Stearic acid, melted 0.946 " " crystallized 0.967-0.969 Beeswax, yellow 0.959-0.962 Ceresin, yellow 0.925-0.928 " white . 0.923-0.924 " very pure white 0.905-0.908 Common resin. . 1.100 DE TERMINA TION OF FAT A CIDS. 265 SPECIFIC GRAVITIES OF SOLID FATS. At 100 (7., compared with water at 15 C. BENEDIKT (1886), from ALLEN (1884) and KONIGS (1883) : A. Fats not containing glycerides of lower fat acids : 1. Vegetable : Theobroma oil 0.857 Palm oil 0.857 Japan wax. 0.873 2. Animal : Lard. 0.861 Tallow (beef or mutton) 0.860 Horse fat. 861 Oleomargarin 0.859 B. Fats containing glycerides of lower (volatile) fat acids : 1. Vegetable : Cocoanut oil 0.863 , Palm-kernel oil . 0.866 2. Animal: Butter fat. . 0.865-0. (7) Determination of the Melting and Congealing Points of Fats. A simple method of finding the melting point is that of the inspection of the fat, taken congealed on the bulb of a ther- mometer, in a beaker of water to which heat is applied, as de- scribed under Stearic acid, a. Sometimes a few drops of the melted fat are taken up in a glass tube of 1 to 3 millimeters internal diameter, bound against the bulb of a thermometer, congealed, and immersed in a beaker of water or other liquid to which heat is gradually applied. The capillarity of the tube influences the movement of the fat, so that the melting point obtained in this way is somewhat higher than that obtained by observation of the fat taken as a coating of the thermometer bulb. Some observers have wider tubes taking a " funnel-tube " of about 2 centimeters diameter in the upper portion and 7 millimeters diameter in the lower portion a few drops of the melted fat being taken and congealed on the side of the wider part, just above the narrowing of the tube. Some authors designate the " beginning of the melting point " as the temperature at which the fat begins to flow down the side of the tube, and " end of the melting point " as that at which it becomes wholly liquid. The sinking point of HEHNEK and ANGELL is the temperature at which a glass bulb of 3.4 sp. gr. and 1 c.c. in volume will sink in the melted fat. The bulb is blown from a piece of glass tub- ing of inch diameter, and is drawn off pear-shaped with a very 266 FA TS AND OILS. tapering end. The bulb should displace as nearly as possible 1 c.c. of water, and should be so weighted by the introduction of mercury as to weigh 3.4 grams. Differences of 0.005 to 0.01 weight have little effect on the results. The following directions for taking "the sinking point" of butter will indicate the method of application to any fat. Of the butter 20 to 30 grams are melted in a beaker over the water-bath, then poured into a test-tube } inch wide and 6 inches long, filling to within two inches of the top. The tube is kept warm until the fat is clarified by the settling of the water, curd, and salt, when the fat is solidified at 15 C. by immersing the tube in water of this temperature. (The cone of depression on the top of the fat serves to indicate its relative fusing point. Pure butter fat shows only a slight depression, while admixtures with fats of high melting points show a considerable hollow cone.) The tube is now placed in about one liter of cold water, in a beaker, the test-tube being se- cured so that the top of the fat is about 1^- inches below the sur- face of the water. Heat is now applied to the beaker by a sand- bath or by asbestos felt, over a lamp. The surface of the fat is made'level, and the weighted bulb placed thereon. The water is stirred from time to time. A thermometer is placed in the water, with the bulb near the surface of the fat, and the temperature read off just as the globular part of the bulb has sunk beneath the fat. Hehner and Angell found the average sinking point of the fat of 24 genuine butters to be 35.5 C. (96 F.), with extremes of 34.3 to 36.3 C. (93.7-97.3 F.) Of the fatty acids of but- ter fat, 40.5 to 42.1 C. Of beef tallow, average, 50.6 C. (48.3 to 53.0 C.); of mutton tallow, 50.9 C. average (50.1 to 51.6 C.) ; of lard, 41.1 to 45.3 C. ; of stearin, 62.8 6 C. ; of cacao but- ter, 34.9 C. ; of palm oil, 39.2 C. To calculate the mean sinking point of a mixture of two fats of known composition, having their respective sinking points (S x and S 2 ), and percentages in the mixture (F x and F 3 ) : ^ lX T^t ( J 2>mpt. rend., 97, 1311; Jour. Gh"m Sor., 46, 532. 3 1883: Analyst, 8, 116; New Bern., 12, 367. 284 FATS AND OILS. gen is influenced largely by presence of fat acids formed by standing open to the air, and that the differences thus due to age and exposure are removed by heating the oil to 204 C. (400 F.) The author holds the value of lubricating oils to be largely dependent on their non-absorption of oxygen. Also, he claims that his method gives the surest detection of cotton-seed oil in admixture with olive oil. He gives the following results in c.c. of oxygen absorbed: Linseed oils Baltic Sea. 191 ; Black Sea, 186 ; Calcutta, 126 ; Bombay, 130 ; American, 156. Cotton- seed oil, refined, 24.6; rape-seed oil, brown, 20; rape-seed oil, colza, 17.6. Olive oil, highest, 8.7; lowest, 8.2. LINSEED OiL. 1 Leinol. Huile de lin. Flachsol. Chiefly the triglyceride of Linoleic Acid (p. 249), C 3 H 5 (C 16 H 2? O 2 ) 3 794. A lixed oil expressed from flaxseed, the seed of Linum usitatis- simum, of which it should form as much as 25 per cent. See Drying and Non-drying Oils, under Fats, p. 281. a. A yellowish or yellow oily liquid, of the sp. gr. about 0.936 (II. S. Ph.), at 15 C. 0.9347 (SCHUBLER), 0.9325 (Sou- CHERE), 0.930-0.935 (ALLEN). Specific gravity of the total fat acids, at 100 C., 0.8599 (ARCHBUTT and ALLEN). Congeals at 16 C. after a few days (GussEROw) ; 27 C. (CHATEAU). Melts at 16 to 20 (GLASSNER). The total fat acids con- geal at 13.3 C. (HUBL) ; melt at 17.0 (HUBL). Boiled Linseed Oil has sp. gr. 0.940-0.941. . Linseed oil has a slight peculiar odor and a bland taste. c. Insoluble in water ; soluble in 5 parts absolute alcohol, in 1.5 parts ether. d. Linseed oil, treated with nitrous acid, does not yield elaidin. Mixed with concentrated sulphuric acid, as directed under Drying and Non-drying Oils, Distinctions between, p. 282, very high numbers are obtained. Treated with iodine, large ab- sorption capacity is found (pp. 258 and 259). For oxygen-ab- sorption see Drying Oils, etc., p. 283. e, f. For Separation and Valuation of Linseed oil, see Dry- ing Oils, p. 281, and Linoleic Acid, p. 249. g t Linseed oil is adulterated with cotton-seed oil, mineral 1 Schaedler: "Technologie der Fette und Oele," 1883, p. 494. Benedikt, " Analyse der Fette," 1886, p. 215. OLIVE OIL. 285 oils, rosin oil, niger-seed oil, rape-seed oil, hemp-seed oil, and fish oils. Mustard, rape, and hemp seeds are gathered with flax-seed. Specific gravity of mineral oils is lighter than of linseed oil, usu- ally from 0.880 to 0.905 ; while resin oil is heavier, 0.96 to 0.99. Non drying oils are indicated by the elaidin test, by not generat- ing the full quota of heat with sulphuric acid, by not absorbing the proper-amount of oxygen, and by lower iodine numbers, ac- cording to directions given under Drying Oils, p. 281. Presence of hydrocarbon or mineral oils is shown by their non-saponifica- tion, sometimes revealed by fluorescence, and sometimes by distil- lation, as specified under Separation of Mineral Oils from Fat Oils, p. 274. Examination for Rosin Oil is directed under the latter, p. 281. Boiled Linseed Oil is prepared by exposure to a high tempe- rature, by which it undergoes oxidation and acquires increased readiness for oxidation. Dryers are added, also, in the " boiling," to promote oxidation by the atmosphere. Manganese and lead oxide, used as dryers, leave traces of these metals in the oil, so that they may be detected in the ash. Boiled linseed oil is frequently adulterated with a very little rosin and with rosin oil. OLTVE OIL. The fixed oil expressed from the fruit of Olea europoe i. Olivenol, Baumol. " Sweet oil." Among the best are .Provence oil and Florence oil. Lucca and Gallipoli oils are good brands. Sicily oil is seldom of best quality. Olive oil is adulterated and substituted by cotton-seed oil, rape oil, poppy oil, sesame oil, peanut oil. a. A pale yellow, or light greenish-yellow, oily liquid, of ,sp. gr. 0.915 to 0.918 (U. S. Ph. and Pli. Germ.) At 15 0., best 0.9178 ; Gallipoli, 0.9196 (CLAKK) ; 0.914 to 0.917 (ALLEN). At 18 C., yellow green 0.9144, dark 0.9199 (STILURELL). Spe- cific gravity of the fat acids, at 100 C., 0.8429-0.8444 (ARCH- BUTT). Congealing point, turbid at 2 C , solid at 6 C. Of the fat acids, congealing point 21.2 C., melting point 26 C. (HUBL) ; congealing point not under 22 C., melting point 26.5 to 28.5 C. (BACH). . Of a nutty, oleaginous, and faintly acrid taste, and nearly without odor. G. Sparingly soluble in alcohol, readily soluble in ether. g. Tests for purity. " If 1 part of olive oil be agitated in a test-tube with 2 parts of a cold mixture prepared from equal 286 FA TS AND OILS. volumes of strong sulphuric acid and of nitric acid of sp. gr. 1.185, and the mixture be set aside for half an hour, the super- natent, oily layer should not have a darker tint than yellowish (a dark color indicating the presence of other fixed oils). If 12 parts of the oil be shaken frequently during two hours with 1 part of a freshly-prepared solution of 6 grams of mercury in 7.5 grams of nitric acid (sp. gr. 1.40), a perfectly solid mass of a pale straw color should result ; and if 1 gram of the oil be shaken for a few seconds with 1 gram of a cold mixture of sulphuric acid (sp. gr. 1.830) and nitric acid (sp. gr. 1.250), and 1 gram of disulphide of carbon, no green or red layer should separate on standing. If 5 drops of the oil are let fall upon a thin layer of sulphuric acid in a flat-bottomed capsule, no brown-red or dark zone should be developed within three minutes at the line of contact of the two liquids (absence of appreciable quantities of other fixed oils of similar properties)." U. S. Ph. " At about 10 C. it begins to grow turbid by crystallization, at C. acquires a salve-like thickness. When 5 grams of the oil are shaken with 15 drops of nitric acid of sp. gr. 1.38, neither the acid nor masses swimming upon it should take a red color. Fifteen parts of olive oil which have been strongly shaken with a mixture of 2 parts water and 3 parts of fuming nitric acid should form a whitish mass, not red nor brown, separating in 1 or 2 hours in a solid mass, while the liquid is scarcely colored." Ph. Germ. " Pale yellow or greenish-yellow, with a very faint, agreeable odor, and a bland, oleaginous taste ; congeals partially at about 36 F. (2.2 C.) " Br. Ph. Specific gravity of mixtures of Olive oil with stated percentages of other oils, at 15 C. (SOUCHERE, 1881, using Lefebre's Oleometer) : Name of the oil. Sp. gr. of the oil admixed. IQperct. 20perct. SO per ct. 40 per ct. 50 per ct. Olive 0.9153 Colza. . 9142 91519 91508 91497 91486 91475 Sesame Cotton- seed. . Walnut 0.9225 0.9230 0.9170 0.91602 0.91607 91547 0.91674 0.91684 91564 0.91741 0.91761 0.91581 0.91818 0.91838 0.9159S 0.91890 0.91915 0.91615 cRSITY OF TURKEY-RED OIL. COTTON-SEED OIL. 287 Congealing and Melting points of Fat Acids of Olive oil with stated percentages of fat acids of other oils (JDACH, 1 1883) : Fat acids from Congealing at Melting at Pure olive oil. . . . Above 22 C. 26 5-28 5 C Olive oil with 20$ cotton-seed oil 28 31 5 Olive oil with 20$ sunflower- seed oil 18 24 Olive oil with 33^$ rape-seed oil 16.5 23 5 Cotton-seed oil, alone 35 38 Castor oil, alone ..... 2 13 TURKEY-RED OIL. A thoroughly non-drying oil suitable for technical ^use in dyeing cotton turkey-red. Grades of olive oil have generally been used. Partly unripe olives, macerated in boiling water before being pressed, yield an oil rich in extractive matter and favorable for this use. In the elaidin test a solid and firm elaidin, of white color, should be obtained. OLIVE-KEKNEL OIL. From the kernel or nut of the olive, by pressure, or extraction with carbon disulphide. Distinguished from olive oil by its dark greenish-brown color and a quite free solubility in alcohol or glacial acetic acid. COTTON-SEED OIL.' Baumwollensamenol or Baumwollenol. Huile de Coton. Oleum gossypii seminis. A fixed oil ex- pressed from the seed of species of Gossypium. The crude oil contains as much as 15 pounds of color substance per ton (LONG- MORE). The color substance, termed gossypin, is soluble in alka- lies, and from alkaline solution precipitated by acids. Treatment with soda lye of five or six per cent, at 60 F. is adopted in puri- fication of the oil, only a small part of which saponifies. The soapy mixture containing the color, when digested with strong soda lye and then neutralized with sulphuric acid, yields a pre- cipitate of the gossypin. 1 Chemiker-Zeitung, 7, 356 ; Zeitsch. anal. Chem., 23, 259 ; Am. Jour. Phar., 55, 354. 2 Production, Uses, and Properties, C S. Munroe, 1885: Am. Chem. Review, 5, 23. Purification, J. Longmore, 1886: Jour. Soc. Chem. Industry. 288 FATS AND OILS. a. The crude oil is a thick, brownish, turbid liquid, which deposits a slimy residue. The clarified oil is clear orange yel- low ; better purified grades light yellow. Fully refined cotton- seed oil is of a very pale straw color. Sp. gr., at 15 C., 0.922-0.930 (ALLEN); 0.9228 (VALENTA) ; at 17 C., 0.923 (SCHEIBE); at 18 C., crude oil 0.9224, refined oil 0.9230, white oil 0.9288 (STILUEELL). Sp. gr. of the fat acids at 100 C., 0.849 (AROHBUTT). Congeals to deposit stearin at 12 C. ; solidifies fully at to 1 C. The fat acids congeal at 30.5 C. (HiiBL), at 35.0 C. (BACH), at 35.5 C. ( YALENTA) ; melt at 35.2 C. (ALLEN) ; begin to melt at 39-40 C., melt wholly at 42-43 C. (BENSEMAN). &. The well-refined oil has only a slight earthy odor, and a bland, perceptibly nutty taste. c. Solubility in glacial acetic acid, according to Valenta, is stated at p. 273, and furnishes a distinction from olive oil. d. Stirred with potassium hydrate solution, crude cotton- seed oil colors blue in the upper layer, becoming violet on expo- sure to the air. The same colors are developed on saponifying with alcoholic potassa, but are hardly made perceptible with the most fully refined oil. When a drop of sulphuric acid is added to a larger quantity of unrefined oil, bright red to brown colora- tion is produced. The test is better made with near equal quan- tities of oil and sulphuric acid of sp.gr. 1.76, gently warming the mixture after observing the first effect. The refined oil re- sponds very slightly to this test. In the elaidin test cotton-seed oil gives elaidin, with reddish-yellow to brownish-yellow colors, these tints being obtained also with nitric acid of sp. gr. 1.42 added in equal volume. Silver nitrate in ether-alcoholic solu- tion is gradually reduced, with dark colors, but t. lis is in a de- gree common to seed oils and olive oil. BECHI (1885) uses a \% solution of silver nitrate in strong alcohol, adding 5 c.c. of this solution to a mixture of 25 c.c. alcohol and 5 c.c. of the oil, and warming to 84 C., when olive oil, he states, does not color if cotton- seed oil be absent. Cotton-seed oil contains about 1.64$ of non- saponifiable matter (ALLEN and THOMPSON, RODIGER). By full saponification and extraction of the dry soap with petroleum benzin (p. 275) a distinction from olive oil is obtained. Cotton-seed oil is a very feebly drying oil. For its identifi- cation, and distinction from olive >il, by this property, tests are made by oxygen-absorption (p. 283), wanning effect of sulphuric acid (p. 282), and the iodine numbers (p. 258). Its separate fat COTTONSEED STEARIN. CASTOR OIL. 289 acids, in the oxygen-absorption test, unlike the entire oil, rank with wholly non-drying oils. The high melting and congealing points of the fat acids of cotton-seed oil distinguish it from most other similar oils (&, and pp. 265, 269). Distinctions between cotton-Seed oil and olive oil are further given under the head of the latter, p. 285. The saponification numbers of olive oil and cotton-seed oil are too near each other to furnish a means of distinction. COTTON SEED STEARIN. Baumwollenstearin. Vegetable Mar- garin. Vegetable Stearin. The residue of cold-pressed cotton- seed oil. A sample examined by MUTER had sp. gr. 0.9115-0.912 at 37.7 C. (100 F.), and gave 95.5$ insoluble fat acids, perfectly soluble in hot absolute alcohol as well as in ether. The melting point was 32.2 C., the melted liquid having a yellow color and odor of cotton-seed oil. It congealed again at about 1 C. A sample examined by Mayer melted at 39 C. CASTOR OIL. Oleum Ricini. Kicinusol. Huile de ricin, de castor. A fixed oil expressed from the seed of the Ricinus communis. See Ricinoleic Acid, of which it is in chief part the glyceride, p. 248. Eiciiiolein is C 3 H 5 (C 18 H 33 3 ) 3 = 931. a. "An almost colorless, transparent, viscid liquid; of sp. gr. 0.950-0.970." U. S. Ph. Sp. gr. at 15 C., 0.9613-0.9736 (VALENTA); at 18 C., 0.9667 (STILURELL) ; at 23 C., 0.964 (DIETERICH). Congeals at 10 to 18 C. Fat acids congeal at 3 C. ; melt at 13 C. (HiiBL). " When cooled it becomes thicker, generally depositing white granules, and at about 18 C. (0.4 F.) it congeals to a yellowish mass." U. S. Ph. b. "Of a bland, afterwards slightly acrid and generally of- fensive taste, and a faint, mild odor." tl. S. Ph. .__ Soluble in an equal weight of alcohol [0.820 at 15 6 C.] and in all proportions of absolute alcohol or glacial acetic acid." U. S. Ph. At 15 C. in 2 parts 90$, and in 4 parts 84$ alcohol. Not soluble in petroleum benzin, kerosene, or paraffin oil, but dissolves about one and a half volumes of kero- sene or paraffin oil. The solubility in alcohol is much varied by temperature. d. Castor oil is to a very slight extent a drying oil. Expo- sure to air causes it to become perceptibly thicker. In the heat- ing effect of sulphuric acid (p. 282), and in the iodine number of 290 FA TS AND OILS. the oil or its acid (p. 258), castor oil stands not far from olive oil. It gives the elaidin reaction. Its saponifieation number is comparatively low (p. 257). g. Impurities and substitutions. The solubilities in alco- hol and in glacial acetic acid, quoted under c from the U. S. Ph., furnish a generally satisfactory means of revealing impurities. The absorption of a little petroleum benzin has been used by Hager for detection of adulterations as follows : One volume of the oil is agitated in a test-tube with 2 volumes of the benzin, and set aside. The lower layer should be increased to from 1.6 to 1.75 of the original volume of the castor oil. In case of adul- teration the lower layer will be proportionally deficient. Under direction of the New York State Board of Health, in 1881, Prof. G. C. Caldwell ' examined 16 samples, of which 9 were considered adulterated, 1 with sesame oil, 4 with cotton-seed oil, 2 with peanut oil, and 2 with cotton-seed oil or peanut oil or both. While giving a caution against dependence upon single tests (not thoroughly subjected to control analyses), Prof. Cald- well advises the legal adoption of test limits. LARD. Adeps. Schweinschmalz. Graisse de pore.' 4 The prepared internal fat of the abdomen of Sus scrofa, purified by washing with water, melting, and straining." a A soft, white, unctuous solid, of sp. gr. about 0.938 (U. S. Ph.) at 15 C.; at 100 C. (water at 15 = 1) 0.861 (Ko- NIGS). Melts at or near 35 C. (U. S. Ph.), 42-48 C. (KoNiGs). At 26 C. melted lard begins to congeal, and during congelation the temperature rises to 30 C. (SCHAEDLER). The fat acids melt at 35 C., and congeal again at 34 C. (MAYER). When rancid, lard acquires a yellowish color. I. Lard when fresh and good has a faint odor free from rancidity, a bland taste, and a neutral reaction. In the air it soon becomes rancid and of an acid reaction. c . It is entirely soluble in ether, petroleum benzin, and disulphide of carbon. ^ e ^ y.__The saponification number of Kottstorfer, for lard, is 195.8 ; 195.3-196.6 (YALENTA). The iodine number of Hubl, 59.0. The per cent, of insoluble fat acids (Hehner's number), 96.15 (WEST-KNIGHTS) ; of non-saponifiable matter, 0.23 (ALLEN and THOMPSON). The per cent, of olein is given by BRACONNOT 1 Analyst, 7, 97; from Sanitary Engineer. LARD. 291 at 62 ; by calculation from the iodine number, 68.4. The melt- ing and congealing points of the fatty acids are named under a. See Fat Acids, Quantitative Determination of, (1) to (9), pp. 250, 274. g. Impurities. Of these the most common is dilution with water, either with or without lime, alum, or other adjuvant, forming " watered lard " ; and those next most probable in this country are " cotton stearin " (p. 289), and mixtures of tallow or tallow stearin with cotton-seed oil. C > <,unut oil has also been used. " Distilled water, boiled with lard, should not acquire an al- kaline reaction (absence of alkalies), nor should a portion be colored blue by iodine (absence of starch). A portion of the water, when 'filtered, acidulated with nitric acid, and treated with test solution of nitrate of silver, should not yield a white preci- pitate soluble in ammonia (absence of common salt). When heated for several hours on the water-bath, under frequent stir- ring, lard should not diminish sensibly in weight (absence of water)." U. S. Ph. " Hot alcohol shaken with the lard, and when cold diluted with an equal part of water, should not affect litmus-papers. When 2 parts of the lard are boiled with 2 parts of (15#) potash solution and 1 part of alcohol, until the mixture becomes clear, and evaporated over the water-bath, the residual soap should dis- solve in 15 parts of warm water on the addition of 10 parts of alcohol." Ph. Germ Dr. MUTER, in 1882, 1 made report to the Public Analysts of a sample of lard adulterated with "cotton stearin," of density of 0.912 as a liquid at 37.8 C. (100 F.) ; yielding 95.5$ fat acids all insoluble, and requiring some time to solidify at 4.4 C. (40 F.) The high density was the obvious indication that it was not lard alone. The sample represented a grade appearing on the market. Alleged adulteration of " prime steam lard" with cotton-seed oil products, in Chicago in 1883, was made the occa- sion of a protracted trial before the Board of Trade of the City of Chicago. 2 1 Analyst, 7, 93: 2 The evidence of a good number of American chemists in this case, with proceedings and findings, was published by the Board of Trade: "McGeoch, Everinghain & Co. vs. Fowler Brothers," pp. 280, 1883, Chicago. Samples of known admixture of lard with tallow, lard with cotton-seed oil, and of pure lard were submitted to four chemists and to one microscopist for analysis. The reports were not in accord with each other, and to a great extent failed of their object. A method of separation by use of alcohol-ether as a solvent was 292 FATS AND OILS. Respecting Microscopical Examination of lard and other fats, Dr. J. H. LONG has contributed a valuable summary, 1 with origi- nal micro-photographs. LAKD OIL. Schmalzol. Speckol. By pressure of lard at about C., the residue being the Solar Stearin or Lard Stearin of the candle industry. Sp. gr. of lard oil, 0.915 (ALLEN). Sa- ponification number, 191-196 (MOORE). TALLOW OIL. Talgol. By pressure of tallow at low tempe- ratures, much lower than are employed for oleomargarin. OLEOMABGARIN. This term has been primarily applied to a product of purified fresh tallow with rejection of a good part of its stearin. The invention of Hippolyte Mege, of Paris, France, prescribed that fresh suet should be immersed in a brine of com- mon salt and sodium sulphite or other addition, then crushed be- tween rollers and washed, and digested at 103 F. (39.4 C.) with sheep's stomachs and calcium biphosphate (U. S. patent, 1873), at animal temperature with infusion of a pig's stomach in acidu- lated water (Br. patent, 1869 ; Bavarian patent), with fresh sheep's stomachs and a very little carbonate of potash in other specifications. By specifications at the 103 F. or at the animal temperature, but practically at 112 F. or the necessary tempe rature, the digested tallow (freed from the cells) melts, and is decanted. It is now allowed to cool to some adjusted tempera- ture (86 to 98 F.), and kept at rest to " crystallize " out " stearin in the form of teats." The decanted liquid is further (or instead of the operation of " crystallization " ) cooled to solidify, and then subjected to pressure, either in a press or in a centrifugal ex- tractor. The liquid product was oleomargarin. 2 much employed, as were also the color tests of Chateau. One witness only states the use of Kottstorfer's method, and no mention, was made of determina- tions by iodine numbers, or by the congealing and melting points of the fatty a.cids after removal of oleic acid. Separation of oleic acid by action of ether on the lead salts was cited by several of the witnesses. Prof. Remsen stated that, in his opinion, " the limit's of variation in the composition of lard " had not been ascertained so as to enable a chemist to determine the question of its adultera- tion. 1 "Some Points in the Micro-Chemistry of Fats." JOHN H. LONG, Sc.D. Chicago Academy of Sciences, 1885. 9 Patents were issued in 1871 in the United States to H. Bradley and to W. K. Peyrous to make the grosser animal fats equal to the best lard, 'for cooking purposes. In 1873, besides the U. S. issue of the Mege patent, a patent was is- sued to E. Q. Parafe for the manufacture of an article designated as " oleo- margarin." Further see F. BONDEL. 1874: " Extract from report of MEGE- BUTTER. 293 The transformation of this into " artificial butter " was under- taken bj Mege through churning with milk, or cows' udders, or other treatment. 1 In this country manufacturers of oleomarga- rin have been practically free from any restrictions of patents, and have conducted the details according to methods governed by the discretion of each producer. Indeed, everywhere the general essentials consist most often in melting at 60 to 65 C. (140 to 149 F.), decanting to clarify, "crystallizing" at 35 C. (95 F.), and pressing at this temperature for "prime margarin" or oleomargarin, the solid residue being known as " prime press- tallow," and used mainly for candles. At the present time lard is extensively treated for a product corresponding to oleomargarin, and used in butter substitutes. In these, also, cotton-seed oil, or fractions of it, sesame oil, and cocoanut oil have been employed in some quarters. Oleomargarin proper, from tallow, has given the following numbers : Sp. gr. at 15 C., 0.924-0.930 (EAGER) ; at 100 C., 0.859 (Konigs). For percentages of stearin in oleomargarin, corresponding to degrees of congealing point of the total fat acids, see table at p. 272. Iodine number, 55.3 (HUBL), 50.0 (MOORE). Ilehner's number (per cent, insoluble fat acids), 95.56. Saponih'cation number of Kottstorfer, 195 to 197.4. Reichert's number, 0.4 to 0.6 (CORNWALL). BUTTER.* The immediate constituents are four : water, curd or casein, salt, and fats. Of these the following percentages occur : Water : proper maximum of good butter, 12$ (HEHNER, WILEY). Of 19 samples reported by the Agricultural Dept. the lowest percentage of water was 7.34 ; the highest, 14.31 ; next highest, 14.06$. Of 49 samples reported by the Board of Health MOURIEZ to the Board of Health of the Department of the Seine on the Product named 'Artificial Butter,'" Amer Chemist, New York, 4, 370. MfiGE-Mou- RIEZ, 1872: Moniteur Scienti,fique, [3], 2, No. 369; Amer. Chem., 3, 231. H. A. MOTT, 1876: "Manufacture of Artificial Butter," Amer. Chem., 7,233. Agriculture of Pennsylvania Reports, 1885: pp. 219-265. "Second Annual Report of the 'New York State Dairy Commissioner," 1886, pp. 190, 312. 1 Further on this subject see TIDY and WIGNER, 1883: Analyst, 8, 113. 2 HEiiNER and ANGELL, 1877: "Butter, its Analysis and Adulterations." A. W. BLYTH, 1882: "Foods," etc., pp. 283-305. Fox and WANKLYN, 1884: "Anal, of Butter," Analyst, 9, 73. BELL, method by sp. gr., 1876: Phar. Jour.. [3], 7, 85. EASTCOURT, sp. gr., 1876: Chem. News, 34, 254. CASSAMAJOR, sp. gr.. 1881: Jour. Am.. Chem. Soc., 3, 81. HAGF.R, odor test of the burning fat, 1880: Zeitsch. anal. Chem., 19, 238. WILEY, 1883-85: Reports of Department of Agriculture at Washington. Also, New York State Dairy Commissioner's Report for 1886. See citations under Butter Fat, p. 298; under the several processes of estimation of fats, pp. 256, 257, 260, 269. SCHEFFER, 1886: Pharm. Rundschau, 4, 248. 294 FATS AND OILS. of Mass., 1885, the maximum was 12.16 per cent. Curd: from 0.5 to 1.2$ (WILEY) ; averaging 2.2$ (HEHNER and ANGELL). Salt : average, 2.5$ (HEHNER). Of 22 analyses of WILEY, from 1.9 to 4.4$. "Should never exceed 8$ " (H. and A.) Fats: the remainder, in the 30 analyses of Hehner and A., highest, 90.2$ ; lowest, 76.4$. A more minute account of constituents is undertaken by KONIG,' for average of 123 samples : fat, 83.27$; water, 14.49; casein, 0.71; milk sugar, 0.58; salts, 0.95 ; of the dry substance, the fat being 97.34$. The fats con- tain traces of color matter, lecithin, and cholesterin, besides the glycerides. Water. The estimation of the water is accomplished by dry- ing a weighed quantity of 3 or 4 grams in the air-bath at a tem- perature not above 110 C., using a tared porcelain or platinum evaporating dish of about 40 c.c. capacity. Traces of free vola- tile acid may be present, and will be driven off. The dish is shaken from time to time. When a nearly constant weight is reached the difference of weight is taken. Absolute alcohol may be added to favor the removal of the last portions. Other opera- tors add a weighed quantity of pure dried sand. -Fat. The residue in the dish is melted, and treated with por- tions of about 10 c.c. of ether (or petroleum benzin), decanting the clear ethereal solution, filtered if necessary, into a weighed beaker. The filter should be dried and weighed in a filter-tube or pair of watch-glasses. The treatment is continued until a portion of the ethereal solution, evaporated on a glass slide, ceases to leave an oily residue. On evaporation or distillation of the ether or benzin the weight of the fat is obtained, or this may be afterward calculated from the difference between the weight of the butter and the sum of the weights of the water and the curd with the salt. According to WILEY, the best method of separating the curd is to dry the butter in a dish (without sand) with a small stirring-rod, at 105 C., adding a little absolute alco- hol. The dried residue when cold (and after weighing for water- loss) is treated with ether or petroleum benzin, and filtered and washed through a Gooch crucible, using an ether-wash-bottle. The crucible is dried at 105 C. The weight of residue, dimin- ished by the amount of salt determined as sodium chloride by volumetric estimation, equals the weight of curd. The fat may also be separated by keeping the butter melted for some time in a tube until it rises in a perfectly clear liquid. In a graduated tube the volume of fat may be taken for its approximate estimation. 1 "Die mensch lichen Nahrungs- und Genussmittel," 1883, ii. 279. BUTTER. 295 Curd and Salt. The residue insoluble in ether is dried and weighed, adding the increased weight of the filter after drying it with its contents arid weighing in the filter-tube, the sum repre- senting the curd plus the salt. The residue is now burned, with the filter, in the dish, to a white ash, weighed as the salt. If for any reason it be desirable, the " salt " may be examined and its so- dium chloride estimated volumetrically. Also the casein may be estimated directly, with exclusion of milk sugar, by washing the residue insoluble in ether with water acidulated with acetic acid, drying, and weighing. Wiley estimates the casein by moist combustion with alkaline permanganate, nesslerizing the distil- late, and calculating from the nitrogen. Separation of artificial coloring matters of butter fats and oils (MARTIN, 1885). To 5 grams of the dry butter fat, or any dry fat, add 25 c.c. of carbon disulphide, and shake gently until the solution is complete. Now add 25 c.c. of water, made slightly alkaline with caustic soda or potash, and shake again gently. The alkaline water will dissolve out the coloring matter. This can be separated out and determined qualitatively by the spectro- scope or other means, and quantitatively by making up a stand- ard solution of annato, or whatever the color may be, and apply- ing the colorometric method. If the color is slight a quantity larger than 5 grams should be taken. Alcohol may be applied to melted butter to extract artificial color. Besides annato, cur- cuma and saffron are employed (MUNICIPAL LABORATORY OF PARIS); also carrot extract and marigold. Carrot extract, in carbon disul- Ehide solution, is not dissolved out by alcohol on shaking with the itter, until a drop of ferric chloride dilute solution is added, when, after shaking and standing, the alcohol extracts the carrot color completely, and if no other color be present the carbon di- sulphide solution becomes colorless (R. W. MOORE). An account of colors proposed or asserted to have been added to butters is given in the report of the N". Y. State Dairy Com- missioner for 1886. The natural color of grass-fed butter, to which the name " lactochrome " has been applied, is bleached by exposure to air and sunlight, and disappears upon about eight hours' exposure to direct sunlight in a layer of 0.5 centimeters thickness (SOXHLET). Mancidity of Sutlers. KOTTSTORFER 1 estimates the free acid of butters as a measure of their rancidity, using the method of simple titration by alcoholic potash in an ether solution of the fat, as adopted for fats in general by GEISSLER, with details given 1 1879: Zeitsch. anal. Chem., 18, 436; Jour. Chem. Soc., 36, 1069. 296 FATS AND OILS. on p. 277. Three to ten grams of clear butter fat, prepared by melting and filtering (as directed under Butter Fat, p. 299), are- weighed into a flask of about 50 c.c. capacity, treated with ether (freed from acidity as directed on p. 278) enough to dissolve it, phenol-phthalein added, the acid of the mixture titrated with the alcoholic potash, and the value of the latter taken by titration with decinormal acid, just as directed on p. 278. The degree of acidity is equal to the number of c.c. of normal solution of alkali required to neutralize the acid in 100 grams of fat. Also, c.c. normal solution X 0.088 = grams butyric acid. And, taking 100 grams of fat, c.c. normal solution X 0.088 = per cent, of free fat acids, estimated as butyric acid. The percentage in the butter fat X .85 (or the found proportion of fat in the butter) = percentage of the butter. The author reports the acidity of the fats of 24 butters. Of these 19 did not overgo 8 degrees acidity of the fat, and the average acidity of the 19 was 4. Of the five above 8 one reached 41.6. The author gives the opinion that for good butter the fat should not exceed 8 degrees of acidity (0.704$ butyric acid). The average of good butters, 4 acidity, corresponds to 0.372 per cent, of butyric acid in -the fat, or about 0.32 per cent, in the entire butter. Detection of Foreign Fats l>y their relation to Solvents. Tests of butter fat by solvents have been proposed as follows : HOORN (1872) used petroleum benzin of sp. gr. 0.69 at 15 C. and boiling at 80-110C. FILSINGER (1880) applied a mixture of 4 vols. ether of sp. gr. 0.725 with 1 vol. alcohol of sp. gr. 0.805, at 18-19 C., for 12 hours. W. G. CROOK (1880) employs, as a "first test " of butter fat, carbolic acid (1 Ib. Cal vert's No. 2 with 2 f. oz. water). Of the purified fat 10 grains (0.648 gram) are treated in a (graduated) test tube with 30 minims (1.5 c.c.) of the carbolic acid, at about 66 C. (150 F.), with agitation. After some time pure butter forms a perfect solution. Fat of tallow (beef or mutton) or lard separates in a defined layer, as does olive oil. The percent- ages of volume of the mixture, taken by the lower (heavier) layer, are given for each of these fats, from 44 to 50 per cent. 1 On cooling, more or less precipitate occurs in the upper layer. So small a proportion of lard as five per cent, did not appear in a lower layer, but after 24 hours gave a crystalline turbidity un- like that of butter fat. A method of distinction between true butter fat and the meat 1 Analyst, 4. Ill; Zeitsch. anal. Chem.(with notes by LENZ), 19, 369. BUTTER. 297 fats by solubility in a mixture of atnyl alcohol and ether is proposed by E. ScHEFFER. 1 Of rectified amyl alcohol 40 volumes are mixed with 60 volumes of ether of sp. gr. 0.725. One gram of the clear butter fat is treated, in a test-tube of 12 c.c. capacity, with 3 c.c. of the solvent mixture. After tightly corking the tube is digested, with shaking, in a water- bath gradually raised from 18 to 28 C. If the butter fat is pure a clear solution is obtained. If the solution be incomplete more of the solvent is added from a burette. For 1 gram of unmixed lard 16 c.c. of solvent is required ; for 1 gram of tallow, 50 c.c. ; for one grain of pure stearin, 550 c.c. of the solvent. A mixture of 10$ lard fat with butter fat took 3.9 c.c. ; 20$ lard fat, 4.8 c.c. ; 40$ lard fat, 6.5 c.c. ; 90$ lard fat, 14.4 c.c. The theory of the test is simply that there is more tristearin in meat fats than in butter fat, although not all the stearic aciM of the latter is held in tri- stearin. Cohesion figures 2 are obtained in " BLYTH'S Pattern Pro- cess" (1880), fully detailed by the author in his work on " Foods " (1 882), giving distinctions between butter fat and various other fats. Viscosity of Butter Soaps. The recent report of S. M. BABCOCK 3 shows that oleomargarin soaps are far more viscous in solution than butter soaps. From the differences given, in tabular comparison, it appears probable that when this method of examination shall have been perfected, it will serve to distin- guish true butters from mixtures of foreign fats, even when the percentage of adulteration is as low as five. The author states that " there is little doubt that by making the solution more al- kaline the addition of one per cent, of adulteration to any given sample of butter would be shown." In the determinations the New Viscometer 4 of the author was used. It is of further inte- rest that the insoluble fat acids of butter were found to give soaps much less viscous than those of corresponding mixtures of stearic and palmitic acids from meat fats, from which the inves- tigator was forced to conclude that " it is probable that the acids of butter are isomers of those from other fats." Microscopical Detection of Foreign Fats in Butter. This 1886 : Pharm. Rundschau, 4, 248. Favorable review by H. W. WILEY, 1887: Science, 9, 114. 2 TOMLINSON, 1861-62. CRANE, 1875. "Report of the Chemist to the New York Agricultural Experiment Station, Geneva, N. Y., distributed Jan. 30, 1887. 4 S. M. BABCOCK, 1886: Chemical Section of Am. Asso. Advance. Sci., Buf- falo Meeting. 298 FA TS AND OILS. subject has been recently reported upon quite fully by Dr. THOMAS TAYLOR/ Reports upon Taylor's microscopical method have been made by Prof. H. A. WEBER a and by Prof. H. W. WILEY.* The last-named two observers do not find that mix- tures of butter with tallow and lard products can be distinguished from pure butter by the process of I)r. Taylor. A brief summary of microscopical methods is given under "Optical Methods" in the Report of the N. Y. State Dairy Commissioner for 1886, p. 271. A useful monograph, with cuts from original micro-photographs, was published by Dr. Long in 1885, 4 recounting the examination of butter and other fats. MYLIUS (1879 B ) has also reported on the microscopical ex- amination of butter. An odor-test by the burning of butter in a wick has been proposed by HAGER (1880). A wick is placed in the melted butter, lighted and burned for two or three minutes, and extin- guished. Oleomargarin gives the odor of a tallow candle. Of this test Messrs. WALLER and MARTIN (1886) say : " All fats when heated to decomposition yield vapors of acrolein which smell the same in all cases. That part of the fat volatilized which has suffered only partial decomposition is what is observed, and is at best a very uncertain quantity. Add to this source of error the fact that old samples of butter have naturally a decided tallowy taste and smell, and it will be seen that the odor in any case is a very uncertain test." BUTTER FAT." Glycerides of palmitic (stearic) and oleic 1 " Butter and Fats. To distinguish one fat from another by means of the Microscope." By Thomas Taylor, M.D., Microscopist to the Department of* Agriculture, Washington, D. C. Proceedings of the American Society of Microscopists. Also various papers, beginning in the Quarfirly Microscopical Journal, New York, 1876. A paper in Proceedings Am. Assoc. Adv. Sci., 1885. 2 Chemist Ohio Agricultural Experiment Station. Bulletin No. 13, 1886, March 1. 3 Chemist Depart. Agriculture at Washington, 1886. 4 "Some Points on the Micro-Chemistry of Fats. JOHN H. LONG. 1885. Chicago Academy of Sciences." *Ber. d. chem. Oes., 12, 270. 6 See citations under " Butter," p. 293. Further, R. W. MOORE, "Notes on Kottstorfer's Method," etc., 1884: Chem. News, 50, 268; "Notes on the Hubl Method," 1885 : Am. Chem. Jour., 6, 416. On Reichert's and other methods. 1885: Jour. Am. Chem. Soc., 7, 188; Analyst, 10, 224. HANSSEN and SCHMITT, 1884: Bied. Cent., 1884, 707; Jour. Chem. Soc., 48, 197. A. H. ALLEN, "On Reichert's Method," 1885: Analyst., 10, 103. KOTTSTORFER on Reichert's and other methods compared with his own, 1879: Zeitsch. anal. Chem., iS.435. REICHARDT, 1884: Archiv d. Phar., 222, 99; Zeitsch. anal. Chem., 23, 565; Jour. Chem. Soc., 46, 1219. BUTTER FAT. 299 nds, and conjugated glycerides of these acids with volatile fat ;ids, (C n H 2n O 2 ), chiefly butyric acid. 1 acids, acids, Preparation from butter, for analysis. The butter is pre- served in the melted state, on the water-bath, with slight shaking or stirring at intervals, until the fat rises in a layer nearly clear. A dry "filter is prepared in a hot funnel, in a warm place, and the nearly clear fat poured on it. The filtrate must be perfectly clear, and not lose weight on the water-bath. For determination of specific gravity the butter is melted at 50 to 60 C., in no case reaching 70 C., and at least 50 c.c. of filtrate obtained. Specific ^gravity, rt, 15 C., 0.926 (CASSAMAJOR), 0.9275 (A. W. BLYTH), 0936-0.940 (HAGER) ; at 37.8 C. (100 F.) (water at same=l), 0.911-0.913 (BELL) ; at 100 C. (water at 15 C.= 1), 0.865-0.868 (KGNiGs) ; at 100 C. (water at 100 C. = 1). 0.901- 0.904 (WOLKENHAAR) ; at 40 C. (water at same = 1), 0.912 (WILEY). Melting Point. Not well defined. Of butter, softens at i>.)..> to 31.1 C., and melts at 24.4 to 37.2 C. (?ARKES and BROWN) ; by the rising of a light glass bulb, mean 33.7, by clearing of the liquid, mean 35.5 C. (HASSALL) ; by the sinking of a weighted bulb, average of 24 samples, 35.5 C. (HEHNER and ANGELL) ; by rising in capillary tubes immersed in water, 31 to 36 C. (HEISCH) ; by the running of a solidified drop of butter, next a thermometer-bulb, on a surface of mercury, 27 to 29 C. (REDWOOD). Of butter fat, 33-36 C. (WILEY). Of the fat acids, 38.0 C. (Hum). Of msoluble fat acids, 39 to 43 C. (WILEY, 1884). Congealing Point. Of biitter fat, 23 to 30 C. (WILEY). Of the fat acids, 35.8 C. (Hum,), '37.5 to 38 C. (MUNICIPAL LABORATORY OF PARIS). Of the insoluble fat acids, 34.5 to 38 C. (WILEY). Per cent, of insoluble fat acids, 87.5 (HEHNER). Milligrams of KOH to saponify 1 gram of fat (KOTTSTORFER), 227. Number of c.c. of T ^ potash solution to neutralize the distilled fat acids 1 The simple glyceride tributyrin does not appear to be present in butters. Conjugated glycerides, such as C3H 5 (Ci6H 31 02)2(C4H702), are inferred to be the sources of the butyric acid of saponification. Mr. HEHNER, however, presents another view : "Both the dipalmitate-monobutyrate and the dioleate-mono- butyrate would yield less insoluble acids than are found in practice, the former 80.2 and the latter 84.6 per cent. But a mixture of compound ethers such as would be obtained by substituting in the Irtpalmitate or ^rioleate of glyceryl one atom of the acid radical by the radical of butyric acid would very approximately yield such proportions of insoluble and soluble fat acids as are actually found." It is to be observed that the question is complicated by the presence of volatile fat acids of larger molecular weights than butyric acid. At all events, the vola- tile fat acids obtained from 100 parts of butter fat average about 6 parts. 300 FATS AND OILS. from 25 gram of fat (REICHERT), 14.0; not less than 13.0 (MEISSL). Iodine number of HUBL, 26.0 to 35.1 ; fat from very old butter, 19.5 (MOORE). Butter Substitutes. Oleomargarin is described at p. 292, with some description of treatment adopted to give it a sensible resemblance to butter. At present prepared lard fat (p. 290) is used as much or more than prepared tallow fat. Frequently a vegetable oil is used with either lard fat or oleomargarin proper (tallow fat). The vegetable oil most used is cotton-seed oil (p. 287) ; after which are to be named sesame oil and cocoanut oil. Of these substitutes two animal fats and three vegetable fats only cocoanut oil approaches in composition to butter fat. It must be remembered that the fats and combinations of fats- presented as substitutes for butter are subject to constant change. The Report of the Dairy Commissioner of the State of New York for 1886 says " the only materials used, according to the statements of the manufacturers, are oleomargarin (' oleo-oil \ lard, cotton-seed oil, sesame oil, and annatto." The term " but- terine" has been more commonly applied to the mixture of deodorized lard and butter prepared by churning with milk. " Suine " is a term applied to a grade of butterine with very large proportions of lard. The work last mentioned describes, besides the oils just named, those of peanut (ground-nut), ben, mustard, colza, rape, cameline, cocoanut, cocoa, palm, cacao, and bone. Principal Chemical Methods of Estimation of Butter-Fat. (1) Parts Insoluble Fat Acids in 100 parts Fat. Hehner's- number, pp. 250, 256. (2) C.c. -j- alkali for Volatile Fat Acids in 2.5 grams Fat. Reichert's number, p. 253. By Meissl's method, p. 253. (3) Milligrams KOH to saponify 1 gram of the Fat. Kotts- torfer's number, pp. 254, 257. PERKINS'S combination plan, p. 255. As a single estimation, that denoted by Reichert's number (probably with Meissl's manipulation) is here unhesitatingly re- commended in preference to any other. But Hehner's number is of nearly an equal value, and next is ranked the number of Kottstorfer, the latter being of the three the most easily ob- tained. Respecting (4) indications by specific gravity, see p. 261. (5) Hiibl's iodine numbers, p. 258. (6) The melting and congeal- ing points of butter substitutes may or may not diifer from that of true butter fat. Respecting the obtaining and applying of data of melting and congealing points, " the sinking point," and BUTTER FAT. 301 " the point of clearance," see p. 265. (7) The percentage of casein, estimated by moist combustion for nitrogen (WILEY) {p. 295), may serve as an aid in establishing a conclusion, though it is to be remembered that a small percentage of poor butter may introduce as large a proportion of nitrogen as would be found in certain samples of the best butter. Interpretation of results. (1) Helmets number. Hehner gives as extremes for true butter fats 86.6 and 88.5 per cent, of insoluble fat acids. If "lower than 88 per cent., the butter must be declared genuine" ; if "higher than 88.5 per cent., we conclude that adulteration has taken place " ; while, " in case sophistication is proved beyond a doubt, we base the calculation upon a lower figure, namely, 87.5." Taking, then, 87.5 as the percentage of insoluble fat acids in true butter, and taking 95.5 as the percentage of the same in fats used as adulterants (p. 256), we have 8 as the difference of Hehner's number due to the sub- stitution of foreign fat for butter. If now a = parts of insoluble fat acids ] In 100 parts A = u foreign fat I , , , B= " butter fat t the tat C = " entire butter equal to the butter fat J D = " true butter in 100 parts of the entire butter ana- lyzed, 8 : ^87.5:: 100 : A. Or, A = 12.5 (0 87.5). B = 100 A. 85 1 : 100 ::B : C. Or, C = 1.1765 B. A + C: 100::C: D. Or, D = ^r.. .A. j vv Of course the factors 87.5, 95.5, and 85 are subject to chemical estimations of the per cent, of insoluble fat acids in butter fat and in adulterating fats, and the per cent, of butter fat in true butter. The per cent, of insoluble fat acids is itself the best statement of results by Hehner's method, and it should be given, for information of those who can understand it, while the calcu- lated per cent, of true butter is given only when required, and may be accompanied with a statement of the conditions on which it is based. The conversion of the percentage of butter fat into that of entire butter is more properly made by use of the actual percentage of total fat found in the butter as sold, instead of the general average figure, 85, above assumed. But even with the use of this factor from the butter in question, there remains the 'Seep. 294. 302 FA TS AND OILS. uncertainty as to how much of the water in the sample was in- troduced as a part of the true-butter fraction, and how much was introduced as a part of the oleomargarin fraction, or was due to manipulation of the mixture. Therefore the opinion is here given that the simple figure known as Hehner's number is the best expression of results by Hehner's method (See the follow- ing corresponding discussion of interpretation of Kottstorfer's number, p. 304). A table of Hehner's numbers of the principal fats concerned in butter adulteration is given on p. 256. Of 29 true butters reported upon by the Department of Agriculture at Washington, 1 live gave between 88.5 and 89.0 per cent, of inso- luble fat acids, three Alderneys gave from 89.0 to 89.26 per cent., and one, an Alderney, gave 89.89 per cent. The Food Analyst of the Pennsylvania Board of Agriculture, Prof. C. B. Cochran, found the extremes of fixed fat acids from fat of 25 genuine but- ters to be 86.T to 87.7 per cent. 2 Rancid Butters give nearly the same percentages as fresh butters (FLEISCHMANN and YIETH), the slight differences being in the direction of an increase. (2) Interpretation of results by ReicJierfs Estimation.* The c.c. of decinormal alkali to neutralize the volatile acids of 2.5 grams of fat. Each c.c. decinormal alkali indicates 0.0088 gram of butyric acid; and 0.0088 gram butyric acid in 2.5 grams of fat is equal to 0.00352 gram butyric acid in 1 gram of fat, or 0.352 in 100 grams of fat. Then, Reichert's number X 0.352 = per cent, of volatile acids (as butyric acid) in the fat ; and per cent, butyric acid -^- 0.352 = Reichert's number. Reichert found true butters to give numbers from 13.55 to 14.45, average 14.0, and declared any butter giving less than 12.0 c.c. must be adulterated. Dr. G. C. CALDWELL reported to New York State Board of Health estimations of 27 samples of butter yielding Reichert's numbers from 12. 7 to 15.5. Messrs. WALLER and MARTIN (Report New York State Dairy Commis- sioner, 1886) obtain, from 26 American butters, on first 50 c.c. of distillate, numbers of Reichert's method from 12.2 to 16.3 as ex- tremes. They also carried eight additional distillates, in exten- sion of Meissl's plan, by which they compute that only from 75 to 85 per cent, of the total volatile acid comes over in the first 50 c.c. Prof. C. B. Cochran, West Chester, Pa., Food Inspector of 1 1884: p. 63, Report of the Chemist, H. W. WILEY. 2 Unpublished report communicated to the author. 3 Directions for estimation and bibliography, p. 253. BUTTER FAT. 303 the Pennsylvania Board of Agriculture, 1 has found the extreme minimum of the Reichert's numbers of known genuine butters to be 12.5 (c.c. of tenth-normal sol. for 2| grams fat) ; and this che- mist holds that Reichert's number 11.5 is the proper minimum limit to govern an analyst in condemning butters inspected by law. He has finally come to rely almost exclusively upon the Reichert's numbers. In Reichert's own analyses lard gave 0.30 ; raw tallow, 0.25 ; rape oil, 0. 25 ; oleomargarin butter, 0.95. Cocoanut oil gave 3.70. Reichert proposes this formula for calculation of per cent. of true butter fat in an admixture of fats : (Reichert's number 0.30) X 7.30 = percentage of true butter fat. The probable error = 0.24 X (Reichert's number 0.30). MEISSL, using his modification (p. 253), places the minimum limit of the Reichert number at 13. R. W. MOORE (1885 3 ) re- ports the following tabulated comparisons of chemical data for the distinction of butter from its substitutes, with discussion of the various methods, and recommends Reichert's method, espe- cially when cocoanut oil is in question. JELubl's number gives the percentage of iodine taken (p. 258) : ._...-; ' Numbers of HEHNEB. KOTTSTOR- FER. HUBL. REICHERT. Butter, samples j 86.01 \ 86.49 ( 95.56 1 89.50 227.0 224.0 197.4 195.0 227.5 19.5 38.0 50.0 50.0 35.4 13.25 13.1. 0.6 0.4 8.7 Oleomargarin, samples. . . Butter, 50$ . . Oleomargarin, 27.5$. Cocoanut oil, 22.5$. The specific gravity of the cocoanut oil used was 0.9167 at 37.7 C., " which is sufficiently high to bring the mixtures above the sp. gr. of 0.911, which is that of butter." WALLER and MARTIN (1886) found cocoanut oil to give a Reichert's number of from 2.7 to 3.7. 1 Unpublished communication to the author. 2 Jour. Am. Chem. Soc., 7, 188; Analyst, 10, 224. 304 FA TS AND OILS. MEISSL (1879) found that soft butters yield higher propor- tions of volatile acids than hard butters. Butter oil gives higher numbers in Reichert's method than entire butter. Rancidity reduces the quantity of the volatile acids of butter. The quantity of volatile fat acids by Reicherfs method is by no means identical with the quantity of soluble fat acids, though the latter should include the former. It will be observed that insoluble fat acids are filtered out of the distillate, if obtained in it, in Reichert's operation. A good number of the analysts of butter practise the estimation of its soluble fat acids by titration of the filtrate and washings from the insoluble fat acids. With- out doubt these results have value. Along with Hehner's method they are easily obtained in a combination process. Di- viding per cent, butyric acid by 0.352, the quotient may be com- pared with Reichert's number. It is believed, however, that Reichert's estimation of volatile acida has greater constancy than an estimatio'n of the soluble fat acids, and therefore the combi- nation plan of PERKINS (p. 255) is given in this work instead of processes including estimation of soluble acids, without distil- lation. The per cent, of soluble fat acids in butter averages at least 5.5, and, according to most authorities, should not fall below 5. Messrs. WAIXER and MAKTIN (1886) obtained from 26 American butters (genuine) the extremes of from 4.49 to 7.25$ volatile acids, as butyric acid, six or seven washings with hot water being taken for the total filtrate titrated (p. 251). It may be remarked that the percentage of total volatile fat acids, from the same but- ters, show less variation, the extremes standing 5.52 and 6.87, as butyric acid, these being obtained by prolonged distillation (beyond the 50 c.c. distillate of Reiehert). (3) Interpretation of Kottstorfer 's number, the milligrams of KOH neutralized in saponifying 1 gram of fat : a measure of the saturating power of the total fat acids. Directions for the estimation, p. 254 ; Table of numbers for Fats and Oils, p. 257. Bibliography, pp. 254, 298. In Kottstorfer's 's conclusion (1879) a number not lower than 221.5 indicates unadulterated butter, this being the lowest limit of true butter. The highest limit he placed at 233, and the ave- rage 227. For oleomargarin the number 195.5 was taken as the average, and for lard the same. If a number (71) be lower than 221 5, the percentage of oleomargarin (x) is found by the for- mula, x = (227 n) 3.17. That is, if the limit number be overpassed by any butter in question, its amount of adulteration is judged by comparison BUTTER FAT. 305 with the average number of true butter a plan corresponding to that followed under Hehner's method. Then we have as data for calculating the percentage (x) of adulterating fat, from Kotts- torfer's number (n) for any mixture of fats : 227 195.5 = 31.5) : (227 n):: 100 : x And The average number of any adulterating fat in question is to be substituted for 195.5 ; and the number 227 is to be held subject to correction as the average number for true butter. The difference 31.5 may be varied by 5.75 in cases of extreme composition of true butter fat, causing 18$ difference in the interpretation. Rancid butters gave Kottstorfer a number, for the fat, 1.5 lower than fresh butter. Mr. WIGNEB, in 1879, stated that "any butter fat which re- quires near 22.6$ KOH for saponih'cation [number 226], as de- termined by the titration process, may be safely passed as gen- uine ; but any lower result should be checked by a further analysis:" (4) Specific Gravity as a means of distinguishing the Fat of Butter from that of its Substitutes. 1 Specific-gravity list, p. 299. Taken by Mr. Bell as a liquid at 100 F. (water at same = 1), using a specific-gravity bottle. By Mr. Cassamajor, as a solid, at 15 C., floated in alcohol of known density. By Mr. "Wigner, as a liquid, at temperatures adjusted (water at 60 F. = 1) by the suspension of specific-gravity bulbs, using data furnished. By Mr. Estcourt, as a liquid, at near the boiling point of water, by use of the Westphal balance. According to Mr. Bell, butter fat (not rancid) rarely falls be- low 0.910 (at 100 F., water at same), the usual range being 0.911 -0.913, and the fat of rancid butter sometimes falling in density to 0.908. The fats of tallow and lard, 0.902.8 to 0.904.6. Cassamajor found that true butter fat, congealing in the al- cohol from melted drops, was held in equilibrium at 15 C. by alcohol of specific gravity 0.926 [15.6 C.] or 53.7 per cent. [volume]. Oleomargarin, treated in the same way, was held in equilibrium at 15 C. by alcohol of sp. gr 0.915, or of 59.2 per 1 J. BELL, 1876: Phar. Jour. Trans., [3], 7, 85. WIGNER, 1876: Analyst, i, 145. CASSAMAJOR, 1881: Jour. Am. Chem. Soc., 3, 83; Chem. News, 44, 309; Jour. Chem. Soc., 42, 341. HEHNER and ANQELL, 1877: "Butter," pp. 76-86. BENEDIKT, 1886: < ' Analyse der Fette," p. 263. 306 FATS AND OILS. cent, strength [by vol.] Quoting, also, the experiments of LETTNE and HARBURET,' Mr. Cassamajor proposed to estimate proportions of oleomargarin, in mixture with butter fat, by a scale of graded strengths of alcohol between 53.7$ and 59.2, cal- culatingon the basis of 5.5 alcoholic percentage for total differ- ence between the two fatty bodies Cocoanut oil has sp.gr. 0.916T at 37.7 C. (100 F.), about 0.0037 above the highest density of butter fat, so that mixtures of oleomargarin and cocoanut oil could easily give the specific gravity of butter fat. Rancid butter fat approaches in specific gravity to the oleomargarin fats. 2 Specific Gravities of Fats and Oils are given in Tables pp. 262, 265. C. B. CocHRAN 3 has used a glass bulb displacing 1 c.c. and of sp. gr. 3.4, for the " sinking point " temperature, and has com- pared the data so obtained with figures of sp. gr. at 100 F. (water at same 1), with results, for spurious butters, as follows : Specific gravity, 905.97 to 911.89. Sinking point, 92.5 F. to 99 F. The Iodine Numbers of HUBL, of frequent application in the analysis of Fats in general, have a very limited application in the analysis of butter, so far as shown for any adulterations hitherto made. Directions for the estimation, p. 258 ; tables, p. 259. Scope of Chemical Analyses of Butter and forms of Certifi- cates, as required of Public Analysts.* The following form of report is used (in 1886) as a tag- record by the Inspector of Foods of the city of Boston. Mass., under the regulations of the city and the laws of the State : " BUTTER : Date, ; Time, A.M. - - P.M. If a store : Proprietor's name, ; No. ; Street, ; sold by ; price paid, ; quantity, Ib. ; wholesalers name, ; jjrice paid ditto, ; District. . If a wagon : Proprie- tor's name, ; name on wagon, ; driver in charge, ; locality, . Butter, . Oleomargarin, . 1 1881: Municipal Laboratory of Paris: Moniteur Scientifique. 3 Further on the effects of Rancidity, JONES, 1879: Analyst, 4, 39. 3 Food Inspector, Penn. Board of Agriculture, in unpublished communica- tion made (1886) to the author. 4 DEPARTMENT OF AGRICULTURE AT WASHINGTON, Reports for 1884, p. 55. Mass. State Board of Health, etc., Reports for 1884, pp. 97, 118; 1885, p. 132. BUTTER. 307 .DULUJimtJ, - . Jlllltai/lUll UUbl/Cl, - . Y HCL11C1 UJiUJVCU piopcrlj, . Analysis .No. F-it Cui'fls - A sli W-itor Tricril nl \\t \nif\a JJ The report of tlie State Board of Health, etc. , of the State of Massachusetts for 1885 gives a list of samples of butter reported upon, with items : " Inspector's number, price per pound, per cent, insoluble fatty acids, remarks." " The highest limit of in- soluble fatty acids in genuine butter fat 90 per cent. has been taken as the dividing line between the genuine and the artificial product." l 1 The Laws of Massachusetts in relation to the Sale and Inspection of Hutler, Oleomargarin, Cheese, etc. [Sections 17, 18, 19, 20, and 21 of Chap. 56 of the Public Statutes, as amended by Chap. 310 of the Acts of 1884, and Chap. 352, Acts of 1885 ] SECTION 1 7. Whoever, by himself or his agents, sells, exposes for sale, or has in his possession with intent to sell, any article, substance, or compound made in imitation or semblance of butter or as a substitute for butter, and not made exclusively and wholly of milk or cream, or containing any fats, oils, or grease not produced from milk or cream, shall have the words il Imitation But- ter," or, if such substitute is the compound known as " Oleomargarin," then the word " Oleomargarin," or, if it is known as " Butterine," then the word " But- terine." stamped, labelled, or marked, in printed letters of plain, uncondensed Gothic type not less than one-half inch in length, so that said words cannot be easily defaced, upon the top and side of every tub, firkin, box, or package con- taining any of said article, substance, or compound. And in cases of retail sales of any of said article, substance, or compound not in the original packages, the seller shall, by himself or his agents, attach to each package so sold, and shall deliver therewith to the purchaser, a label or wrapper bearing in a conspicuous place upon the outside of the package the words "Imitation Butter," "Oleo- margarin," or " Butterine," as the article may be, in printed letters of plain, un- condensed Gothic type not less than one-half inch in length. SEC. 18. Whoever, by himself or his agents, sells, exposes for sale, or has in his possession with intent to sell, any article, substance, or compound made in imitation or semblance of cheese or as a substitute for cheese, and not made ex- clusively and wholly of milk or cream, or containing any fats, oils, or grease not produced from milk or cream, shall have the words " Imitation Cheese " stamped, labelled, or marked, in printed letters of plain, uncondensed Gothic type not less than one inch in length, so that said words cannot be easily de- faced, upon the side of every cheese-cloth or band around the same, and upon the top and side of every tub, firkin, box, or package containing any of said ar- ticle, substance, or compound. And in cases of retail sales of any of said arti- cle, substance, or compound not in the original packages, the seller shall, by himself or his agents, attach to each package so sold, and shall deliver there- with to the purchaser, a label or wrapper bearing in a conspicuous place upon the outside of the package the words "Imitation Cheese," in printed letters of plain, uncondensed Gothic type not less than one inch in length. SEC. 19. Whoever sells, exposes for sale, or has in his possession with in- 3 o8 FA TS AND OILS. Under the action of the Dairy Commissioner of New York State some of the analysts of butter (in 1886) report upon print- ed blanks as follows for butters found to be spurious : " Certifi- cate of Analysis of a - sample of.- - 'Butter'; marked ; received iroin - , per - , on - . This sample contains : Animal [vegetable] and Butter Tor total] Fat, ; Curd,- -; Salt (ash), - -; Water at 100 C. } . Analy- sis of the Fat present in the sample : Soluble fatty acids (by distillation) (on a dry basis), ; Insoluble fatty acids, ; Specific Gravity of the dry fat, at 100 F. Titer, - - [Kei- chert's number]. (This sample is composed - - of foreign fat, and is not produced from unadulterated milk, or cream from the same. It contains coloring matter, whereby it is made to re- semble butter, the product of the dairy, and is made in imitation and semblance of butter produced from unadulterated milk, or cream from the same.) " The Food Analyst of the Board of Agriculture of Penn.^ Prof. COCHRAN, cites the following results of official analyses of butters : tent to sell, any article, substance, or compound made in imitation or semblance of butter or cheese, or as a substitute for butter or cheese, except as provided in the two preceding sections, and whoever defaces, erases, cancels, or removes any mark, stamp, brand, label, or wrapper provided for in said sections, or changes the contents of or in any manner shall falsely label, stamp, or mark any box, tub, article, or package marked, stamped, or labelled as aforesaid, with intent to deceive as to the contents of said box, tub, article, or pack- age, shall for every such offence forfeit to the city or town where the offence was committed one hundred dollars, and for a second and each subsequent of- fence two hundred dollars. SEC. 20. Inspectors of milk shall institute complaints for violations of the provisions of the three preceding sections when they have reasonable cause to believe that such provisions have been violated, and on the information of any person who lays before them satisfactory evidence by which to sustain such complaint. Said inspectors may enter all places where butter or cheese is stored or kept for sale, and said inspectors shall also take specimens of sus- pected butter or cheese and cause them to be analyzed or otherwise satisfactori- ly tested, the result of which analysis or test they shall record and preserve as evidence; and a certificate of such result, sworn to by the analyzer, shall be ad- mitted in evidence in all prosecutions under this and three preceding sections. The expense of such analysis or test, not exceeding twenty dollars in any one case, may be included in the costs of such prosecutions. Whoever hinders, ob- structs, or in any way interferes with any inspector, or any agent of an inspec- tor, in the performance of his duty, shall be punished by a fine of fifty dollars for the first offence, and of one hundred dollars for each subsequent offence. SEC. 21. For the purposes of the four preceding sections, the terms "but- ter" and 4< cheese" shall mean the products which are usually known by these names, and are manufactured exclusively from milk or cream, with salt and rennet, and with or without coloring matter. BUTTER. 309 JKeichert's num- ber. Hehner's num- ber. 'Butter" reported to be No. 1 3.1 c.c. 88.9^ Not genuine. 2 4.2 93.45 a 3 3.0 92.9 u a 4 14.0 Genuine. 5. . . 12.6 a 6. 1.6 Not genuine. 7 14.15 Genuine. 8 13.0 u 9 14.15 tt 10 1.5 Not genuine. 11 0.75 a u 12 1.0 tt a 13 11.7 Passed. 14 0.7 Not genuine. 15. 0.85 u u 16. 1.6 u u IT. 1.0 a u 18. . 2.4 a 19. 12.1 Passed. 20.. . 5.2 Not genuine. 21 12.8 Passed. 22 0.6 Not genuine. 23 13.7 Genuine. 24. . , 3.0 Not genuine. 25 12.5 Passed. 26 12.2 u 27 16.3 u 28 1 13.3 u 29. 15.2 it 30.. 15.5 u The Agricultural Department at Washington for 1884, Prof. WILEY, Chemist, reports tabulated analyses, with statements of u No., name, made at, made by, bought at, price, color [to the eye], water, casein, salt, fat, melting point, solidifying point, saturation equivalent (56000 -f- Kottstorfer's number), soluble 'No. 28: Fats, 78 per cent. ; water, 17.38 per cent.; curd, 2.4 per cent. ; ash, 2.21 per cent. 310 FATS AND OILS. acid (in per cent, as butyric, obtained bj titration of filtrate from insoluble fat acids), insoluble acid, melting point insoluble acid, solidifying point insoluble acid, saturation equivalent of insoluble acid " (56000 -f- Kottstorfer's number for the insoluble acids). What is a sufficient chemical analysis of butter f A single faithful estimation, whether of (1) the insoluble fat acids, (2) the soluble acids distilled, or (3) the saponification number, as these estimations are detailed in pp. 250 to 255, may give such a result that no further evidence is needed to prove the butter to be not a genuine one. And the result of an estimation by any one of these established methods may be in itself sufficient to prove that a certain sample does not consist mainly or largely of oleo- margarin. Besides, oleomargarin, lard products, and cotton-seed oil, or any mixture of these three, may be distinguished with certainty from pure butter fat by any one of the methods just named (Hehner's, Reichert's, or Kottstorfer's). Further, any mixture of oleomargarin, or lard product, or cotton-seed oil, or combination of these foreign fats, with a smaller proportion of butter fat of average or nearly average composition, must be clearly revealed not a pure butter fat by the result of a true esti- mation according to Hehner, or Reichert, or Kottstorfer. Adul- teration with the foreign fats above named, when in proportions not less than half, and when with butter fat of about average composition, can invariably be declared as adulterations by analy- sis under one of the three methods here referred to. And what is here stated as true of oleomargarin or prepared tallow fat, and lard fat, and the fat of cotton-seed oil is known to be true of numerous vegetable fats, some of which have been used in butter substitutes, and is true of known fats with a very few excep- tions. When a question of small proportions of foreign fats in mix- ture with large proportions of average butter fat is presented, it is to be understood that there is a limit to the diminution which foreign fat may undergo and still remain capable of detection by one of the estimations above enumerated, or by any mode of analysis. Just what percentage of foreign fat marks such limit it is difficult to state. Granting that the butter fat of the mix- ture be of average composition, the limit must lie in such low percentages of foreign fat as would be of improbable occurrence in adulteration for commercial ends. But when the possibil- ity of admixture with butter fat of exceptional composition is BUTTER. 311 introduced into the question, the limit of quantity of foreign fat capable of detection by chemical estimation rises to a place among percentages which are quite possible among the devices of adul- teration. If the article be rancid a somewhat abnormal compo- sition of butter fat may have been acquired. A rule has prevailed, in the interpretation of results under several methods of analysis, that only when the result stands out- side of the extremes obtained among the results of varying sam- ples of genuine butter fat shall a butter be declared (qualitative- ly) adulterated. But when so declared adulterated a (quantitative) statement of the proportion of the adulteration may be based upon the deviation from the average of results of genuine butter fat. This rule has been discussed at p. 302. Its bearing may be illustrated by an application under Helmer's method, as fol- lows : If the insoluble fat acids be found at 88.0$ of the clear fat the article is declared not adulterated. If found at 88.5$ the ar- ticle is not declared adulterated (on this evidence alone). If found at 88.6$ insoluble fat acids, an adulteration of 13f$ of foreign fat in the total fat is reported under the rule. At the same time, if the extreme limit of 88.5$ insoluble acids in exceptional butter fat be taken as the datum of calculation for the quantitative report on 88.6$, as was taken for the quali- tative verdict on 88.5$ that is, giving the article on trial the benefit of possibilities in both the cases alike from 88.6$ of in- soluble fat acids only 1.43$ of foreign fat in the total fat would be declared. And a logically guarded report could state that, from the evidence of 88.6$ of insoluble fat acids, it appears that an adulteration of foreign fat has been made, in quantity from about 1.5 to about 23 per cent., and probably near 13 per cent. In order to reach a secure conclusion respecting the fact of adulteration in cases of admixture of foreign fats with large pro- portions of butter fat, and in order to give the percentage of foreign fat within limits brought as near each other as possible, more than one of the estimations (Hehner's, Reichert's, Kottstor- fer's) should be made. The three estimations, with determina- tion of the specific gravity of the fat as a fourth, furnish together a body of evidence more trustworthy in cases of difficulty as to the fact of adulteration, and more exact respecting percentages, than can be drawn from any smaller number of determinations. Still other determinations, as those of melting and congealing points, and the quantity of casein, sometimes give additional ad- vantage. In case of doubt every means of investigation should be used. And all important estimations should be obtained in triplicate or duplicate results. 312 FORMIC ACID. FORMIC ACID. Ameisensaure. CH 2 O 2 rz:46. Hydro- gen-carboxyl, H.CO 2 H, the first member of the fatty acid series, C n H 2n+1 CO 2 H. Obtained by distilling the bodies of ants with water. A constituent of the exudate carried with the stings of insects and of stinging nettles. A product of nume- rous organic reactions, including a rapid action of alkalies upon chloral, and a feebler action of alkalies upon chloroform, also the action of potassium upon carbon dioxide and water, or of hot potassium hydrate solution upon carbon monoxide. Prepared by distillation from oxalic acid and glycerin. A common result of destructive distillation. Formic acid, when free and not dilute, is recognized by its odor and irritating effect on the skin (J). Tests of identification are obtained in the color with ferric salt, the precipitates with lead acetates and alcohol, the reduction of silver or mercury, and the generation of carbon monoxide with sulphuric acid (d). /Separations are made by distilling formate with phosphoric acid, and by the insolubility of lead or calcium formate in alcohol (e). Estimations are conducted acidimetrically, by weight of lead formate, and by weight of mercury reduced (f ). Certain im- purities are liberated by holding calcium formate in alcohol (g). a. Formic acid is a colorless liquid, of specific gravity of 1.221 at 20 C., boiling at 100 C., the T7.5 per cent, acid at 107.1 C., and crystallizing, when pure, below C. Metallic formates heated to decomposition do not form a carbonaceous residue. J. The odor of formic acid is pungent, in proportion to the concentration of its aqueous solution, the vapor from the strong acid having a slightly suffocating effect reminding of sulphurous acid. The taste is purely acidulous. The effect is irritating, the strong acid causing burning and itching of the skin, a biting sen- sation of the tongue, and a tingling of the nostrils. c. Formic acid is miscible in all proportions with water and with alcohol. The metallic formates are generally soluble in water and but little soluble in alcohol. The normal metallic formates mostly exhibit a neutral reaction with litmus-paper ; the normal lead formate being neutral, and the basic lead formate al- kaline in reaction. The formates crystallize readily. d. Ferric chloride, in a neutral solution of alkali formate, forms ferric formate, of red color, and precipitated on boiling, a reaction closely resembling that of acetic acid. Normal lead FORMIC ACID. 313 acetate precipitates concentrated solution of an alkali formate, the normal formate of lead being soluble in 65 parts of cold water. By adding to the mixture twice its volume of alcohol the precipitation is much increased. If basic lead acetate solu- tion be saturated with alcohol it serves to precipitate formic acid, free or combined, quite completely, as the basic formate of lead is very little soluble in alcohol of moderate strength. The pre- cipitate of lead formate dissolves freely in hot water, and on cool- ing the solution needle-form crystals of lead formate are obtained, more perfectly after several hours. Silver nitrate solution gives a white, crystalline precipitate of silver formate, only in quite con- centrated solutions. On standing or warming the precipitate blackens by reduction to metallic silver. In more dilute solutions the metallic silver is the lirst form of the precipitate, and best obtained in neutral or feebly acidulous solution, free ammonia being avoided. Reduction by vapor of formic acid is to be adopted if non-volatile reducing agents are liable to be present, and is accomplished by slightly acidulating the mixture with sulphuric acid and immersing the test-tube for some time in boiling water, while a disk of filter-paper previously wetted or crossed with solution of silver nitrate is bound over the mouth of the tube. Mercurous nitrate gives a precipitate of mercurous formate, soluble in 500 parts of water. Reduction to metallic mercury is obtained on standing twenty-four hours, more readily on warming. Concentrated sulphuric acid, at a gentle heat, resolves formic acid into carbon monoxide and water (OH 2 O 2 H 2 O-|-CO). The formic acid or its salt is warmed with about three times its volume of the sulphuric acid. With a considerable quantity the resulting gas may be burned at the mouth of the test-tube, with a blue flame. No carbon dioxide is obtained, carbonates being absent a difference from oxalic acid. Heated with strong alkali, in the air, formate is changed to oxalate. Ethyl formate is de- veloped by distilling a dry formate with about an equal quantity of alcohol and a double quantity of sulphuric acid, undiluted. The ester has a characteristic fragrance, sajd to resemble that of peach-kernel not sharply distinguished from esters of homolo- gous acids as obtained in qualitative tests. e. Separations. Free formic acid may be separated from water and other non-acidulous volatile bodies by neutralizing with fixed alkali and evaporating to dryness on the water-bath. From the residue, or any portion of formate, by adding phosphoric acid and distilling at 100 C. or a little above. If sulphuric acid 3 H FUSEL OIL. be used instead of phosphoric it must be well diluted, and the dilution maintained by adding water from time to time, long dis- tillation being now required. From acetic acid free formic acid may be separated by digesting with enough lead oxide to cause a permanently alkaline reaction, evaporating to dryness, and ex- hausting the residue with alcohol. The filtrate will contain the acetic acid as lead basic salt, and the residue will contain the formic acid in combination, from which it can be recovered by distilling from phosphoric acid or by thorough treatment with hydric sulphide gas and following filtration. Instead of lead oxide, calcium carbonate or magnesium oxide may be employed, recovering the formic acid by distilling from phosphoric acid. f. Quantitative. Free formic acid may be estimated volu- metrically by titrating with alkali, using litmus or phenol-phthal- ein as an indicator. -The lead formate obtained by precipitation with lead normal acetate and alcohol, as directed under d, may be washed with alcohol, dried, and weighed as normal lead for- mate. In a mixture of formic and acetic acids the formic acid is capable of estimation by its reduction of mercury. The mixture is digested some time with an excess of yellow mercuric oxide, the washed residue treated with dilute hydrochloric acid to re- move the remaining oxide, and the metallic mercury gathered, washed, and weighed. HgO + CH 4 O = Hg + CO 2 + H 2 O. Hg : CH 4 O :: 199.7 : 46 :: 1 : 0.2303. g. Impurities of acetic, hydrochloric, nitric, or other acids forming calcium salts soluble in alcohol may be found by digest- ing the acid mixture with excess of calcium carbonate, evaporat- ing to dryness, and treating with alcohol, when the filtrate will contain the calcium salts of the acids mentioned. FUSEL OIL. Fuselol. Huile de pommes de terre (po- tato oil). The sum of the heavy alcohols obtained as a by-pro- duct in the manufacture of ethyl alcohol in its ordinary forms. A portion of higher-boiling distillate received after distillation of commercial alcohol or distilled spirits. A variable body of amyl alcohols with smaller proportions of adjacent alcohols of the C n H 2n 2 O series, as products accompanying ethyl alcohol in the common alcoholic fermentation. Obtained in the fermenta- tion of potatoes, indian corn, marc of grapes, and in smaller quantities by the fermentation of other materials used as sources of sugar. Fusel oils from their several sources differ from each other in composition, but amyl alcohols form by far the larger part of all of them. Acids of the C n H 2n O 2 series are found in FUSEL OIL. 315 fusel oils, where they are formed by oxidation of the alcohols. Ethereal salts occur by action of fusel oil acids upon fusel oil alcohols and upon ethyl alcohol, as a result of " ageing." The alcohols of fermentation, found in fusel oils, are chiefly as follows : Per cent, by vol. Boiling, C. Spec, grav. 27.5 1 Inactive amyl alcohol, iso- butyl carbinol (CH 8 ) 2 . CH . CH a CH 2 OH . 131.4 0.812 13.0 ) 6.0 1 ] 5.0 1 Active amyl alcohol, second- ary-butyl carbinol CH 3 .C 2 H 6 .CH.CH 2 OH.. . Iso- butyl alcohol, propyl car- binol (CH s ) 2 .CH.CH a OH 128 108.4 0.808* 0.802 6.5'(?) Traces . 3.0' 15 1 (ft Normal butyl alcohol CH 3 .CH 2 .CH 2 .CH 2 OH . . Tertiary butyl alcohol (CH 3 ) 3 .COH Normal propyl alcohol CH 3 .CH 2 .CH.,OH Secondary propyl alcohol CH 3 CH(CH 3 )OH 116 84 97.4 82 8 0.810 0.779 0.807 788 A primary hexyl alcohol. ... CeHi 3 OH . 150 The identification of several of the alcohols given above is not well established. The isomerides of different fusel oils are not the same. Rabuteau (loo. cit.) obtained 17 per cent, of pro- ducts boiling above 132 C., and including some amyl alcohols. ORDONNEAU (1886 : JRep. anal. Chem.) obtained from wine bran- dy twenty-five years old the following : normal propyl alcohol, 0.040$ ; normal butyl alcohol, 0.218$ ; normal [!J amyl alcohol, 0.084$; normal hexyl alcohol, 0.0006$; normal heptyl alcohol, 0.0015$ ; propionic, butyric, and caprilic ethers, 0.004$ ; oenan- thic ether, about 0.004$ ; acetic ether, 0.035$ ; ethylaldehyde, 0.003$; acetal, 0.035$; amine bases, 0.004$. The same author states that alcohol from maize, beets, or potatoes contains iso- butyl instead of normal butyl alcohol, and contains amyl and propyl alcohols, and pyridine, probably collidine. Wine yeast (elliptic) produces normal butyl alcohol; beer yeast (globular) no butyl alcohol. Except the uncertain report of secondary propyl alcohol, above quoted, and the traces of tertiary butyl alcohol, the alcohols found in fusel oil are primary, and therefore capable of forming acids without breaking up. 4 1 KABUTEAU, 1878: Compt. rend., 87, 501; Jour. Chem. Soc., 36, 36. 2 LE BEL, 1873: Ber. d. chem. Ges., 6, 1362. 3 FAGET: Liebig's Annalen, 88, 32o normal hexyl alcohol. 4 Of pentyl alcohols, CtHuOH, eight are possible, and seven are known: three primary alcohols, three secondary alcohols, and one tertiary alcohol. There are four butyl alcohols, C 4 HOH, two primary and two secondary, all 3 i6 FUSEL OIL. The fusel oil acids are not found in quantities sufficient for their satisfactory separation. Of ethereal salts, ethyl salts have been mainly found, instead of amyl salts, in fusel oils. Ethyl alcohol is retained in various quantities, limited by present Eng- lish excise law to be below 15 per cent, of commercial fusel oils. a. Fusel oils are received by distillations beginning at 105 to 125 G., and ending at 132 to 137 C. Between 105 and 120 C. the most of the iso-butyl alcohol is obtained ; between 128 and 132 C. the amyl alcohols are mostly distilled. b. In physiological effects the fusel oils have a stifling, harsh, spirituous odor, quite characteristic and subject to differ- ences which reveal to the expert the source of the fusel oil. Even a slight inhalation excites coughing. Objectionable pro- portions of fusel oil in alcoholic liquors are recognized by first evaporating off the ethyl alcohol and obtaining the odor of the warmed residue. For the U. S. Ph. test of " alcohol " it is pre- scribed that " if mixed with its own volume of water and one- fifth its volume of glycerin, a piece of blotting-paper on being wet with the mixture, after the vapor of alcohol has wholly dis- appeared, should give no irritating or foreign odor (fusel oil)." " A little [rectified spirit] rubbed on the back of the hand leaves no unpleasant smell after the spirit has evaporated " (Br. Ph.) In their effects on the system the alcohols of the C b H 2n+2 O se- ries have an intensity which increases with the molecular weight. RABUTEAU (1870), mainly from experiments with frogs, estimated the intensities of effect of ethyl, butyl, and amyl alcohols to be, respectively, as 1, 5, and 15. Dr. B. W. RICHARDSON' states that the action of amyl alcohol is that of butyl alcohol intensified. In the third stage of the action there are pronounced tremors of regular recurrence, reduction of temperature, and profound coma. Recovery requires sometimes two or three days. In recovery the restoration of the temperature is delayed longest. After death from amyl alcohol the blood is excessively venous. c. /Solubilities. The amyl alcohols of fusel oil are said to dissolve in about 40 parts of cold water. According to BAL- known. The two possible propyl alcohols, C 3 H 7 OH, are included in the list above. Of the seventeen hexyl alcohols, isomerides of C 6 H 13 OH, eight are known at present, four being primary. It is evident, upon a very simple ac- quaintance with chemical law, that there can be but one ethyl alcohol, in what- ever mixture it be found. Differences in the physiological effects of various al- coholic beverages are not due to any difference in the ethyl alcohol contained therein. 1 1875 : Cantor Lectures, London. FUSEL OIL. 317 BIANO (1876 ') the inactive amyl alcohol of fusel oil is soluble in about 50 parts of water at 14 C., and less soluble in water at 50 C. Iso-butyl alcohol is soluble in 10 parts of water at 15 C. One part of inactive amyl alcohol takes up about 0.08 parts of water ; one part of iso-butyl alcohol, about 0.15 parts of water. Fusel oil is freely soluble in ether, chloroform, and the other im- miscible solvents of general use. Of the amylsulpJiates of barium, that formed from inactive amyl alcohol is two and a naif times less soluble in water than is that from active amyl alcohol a dif- ference made available for separation of these isomers. d. In the qualitative identification of fusel oil the odor, and the effect of inhalation, are the means in most common use. In testing fusel oil in liquors or commercial alcohol the ethyl alco- hol is made to evaporate before obtaining the odor. The Br. Ph. and U. S. Ph. directions for examination by odor are given under b. Separation by an immiscible solvent, as described un- der , may be adopted preparatory to any qualitative examina- tion. In ordinary analyses of alcoholic liquors or commercial alcohols, the ethyl alcohol has to be separated by careful distilla- tion for the estimation of " strength," and the residue of such distillation, while warm, is to be examined for odor. Concentrated sulphuric acid, on contact with fusel oils, en- ters into formation of amylsulphuric acids, HC5H n SO 4 , present- ing a red color. 2 According to YITALI (loo. cit.\ when the sul- phuric acid is added to a smaller quantity of fusel oil the color is red, growing brown-red on standing and on heating. Equal volumes of the amyl alcohols and the sulphuric acid give a dark and dull red color ; but with an excess of the fusel oil differ- ent tints are obtained, cherry-red, violet, azure-blue, and green, in the order of the increasing excess of fusel oil. When to a drop or two of sulphuric acid, on a white porcelain surface, an equal volume of the fusel oil is first added, and then additional portions of the latter, cherry-red and violet tints are obtained, and at the proportions of five or six volumes of the fusel oil the azure-blue color is reached. The addition of ether to the colored mixture increases the brilliancy of the tints. The test may be applied to the residue left after evaporation of a chloroformic or ethereal solution, obtained by shaking out 1 Ber. d. chem. Ges., 9. 1437. S PELLETAN, 1825: Ann. Ghim. Phys., [2], 30, 221. Amylsulphuric acid of iso-butyl carbinol (inactive amyl alcohol), CAHOURS, 1839: Ann. Chem.Phar., 30, 291. As a color test for fusel oil, further, VJTALI, 1884: Archiv der Phar., [3], 21, 964; Zeitsch. anal. Chem., 23, 426; Analyst, 9, 196. 3 i8 FUSEL OIL. as stated under , and as so applied the test is trustworthy for negative results showing the absence of material proportions of fusel oil. But inasmuch as numerous non-volatile bodies give colors with sulphuric acid, an indication of the presence of fusel oil should be verified, in this test, by applying it to a fractional distillate from the liquor or commercial alcohol under examina- tion a distillate obtained between about 120 and 134 C. For the qualitative use of MARQUARDT'S plan of separating amyl alcohols directions are given below under f. This method is very delicate. The odor of the valeric acid is highly distinc- tive. The reaction of JORISSEN, obtained by mixing with a lit- tle colorless aniline oil and a few drops of sulphuric acid, for a fine red color, depends on the presence of furfurol (aldehyde be- tween furfuryl alcohol and pyromucic acid) present in fusel oils." SEVALLE (1881) determines the presence of fusel oil by turbi- dity of its alcoholic mixture when heated. TRAUBE 2 tests brandy, after dilution to about 20 per cent, of strength, by the height to which the liquid rises in a capillary tube, as compared with a pure spirit of the same strength. The etherification of amyl al- cohols, to form esters of acetic acid, has been included among methods for recognition. The liquid is warmed or distilled with alkali acetate and sulphuric acid. The odor of amyl acetate is that of pears. This odor, however, is quite liable to be covered by that of acetic acid or ethyl acetate, so that caution should be observed in interpretation of the result. e. The separation of fusel oil by distillation gives prac- tically conclusive results, but is certainly not without waste. Sep- aration by immiscible solvents is generally employed. BETELLI* adds to a certain quantity of the commercial alcohol six or seven times its volume of water, and shakes with chloroform enough to make after subsiding a very small layer, which is drawn off and evaporated for test of the residue. With only 5 or 6 c.c. of the alcohol, and 15 to 20 drops of the chloroform, 0.05 per cent, of fusel oil was detected, the final test being that of etherification to amyl acetate (" pear oil " ) by digesting the residue with al- kali acetate and sulphuric acid. Before applying the immiscible solvent it is proper to reduce the concentration of the ethyl 1 FORESTER, 1882: Ber. d. chem. Ges., 15, 230. 2 1886: Ber. d. chem. Ges., 19. 892; Jour. Chem. Soc., 50, 743. Follows a report on relation between capillarity and molecular weight, and on specific capillarity, 1885: Jour, prakt. Chem., [2], 31, 177, 514; Jour. Chem. Soc., 48, 866, 1033. 3 1875: Ber. d. chem. Oes., 8, 72; Zeitsch. anal. Chem., 14, 197. FUSEL OIL. 319 alcohol, either by distilling it off or by adding water. See the directions given below under/*. f. Quantitative. For the estimation of fusel oil the method of MAKQUARDT ' is here given : The fusel oil is separated from a diluted alcoholic liquid by shaking out with chloroform ; the amyl alcohol is oxidized to valeric acid; the acid is taken up by barium carbonate, and the barium salt is estimated, to repre- sent the quantity of amyl alcohol oxidized. Of the alcoholic liquid under examination 150 grams are to be diluted with an equal volume of water, and agitated with 50 c.c. of chloroform (of ascertained purity) for about a quarter of an hour. The aqueous layer is separated and again extracted with 50 c.c. of chloroform for the same length of time. The operation is re peated the third time. The united portions of chloroform are treated, in a strong flask, with a solution of 5 grams dichromate in 30 grams of water, and 2 grams of sulphuric acid, digesting for about six hours with frequent agitation while the flask is well corked. The contents of the flask are then distilled until about 20 c.c. remain, when this residue is diluted by addition of about 80 c.c. of water, and again distilled until only about 5 c.c. remain. The entire distillate is then digested with heat for half an hour with barium carbonate, an erect condenser being employed to return the distillate to the flask. The chloroform is then sepa- rated by distillation, and the aqueous residue concentrated to a volume of 5 c.c. The excess of barium carbonate is filtered out, and the filtrate with the washings evaporated in a weighed dish on a water- bath to dry ness. The residue is weighed, and after- wards dissolved in water and a few drops of nitric acid, and made up with water to 100 c.c., of which 50 c.c. are taken to es- timate the barium, and 50 c.c. to estimate the chlorine (derived from the chloroform by action of the dichromate). The quantity of barium combined with the chlorine is calculated, and deducted from the total barium in the 50 c.c. From the remaining quan- tity of barium the quantity of valeric acid is calculated, and from this the quantity of amyl alcohol. Then BaSO 4 : 2C 5 H 12 O :: 232.8 : 176 ::1 : 0.7560. And the weight of barium sulphate X 0.756 = the weight of amyl alcohol. For exact results the chloroform used is to be purified by subjecting it to the operation as described treatment with solu- tion of dichromate and sulphuric acid, distillation, digestion of 'L. MARQUARDT, 1882: Ber. d. chem. Ges., 15, 1370, 1661; Analyst. 8, 106; Jour. Soc. Chem. Ind., I, 331, 377; Jour. Chem. Soc., 42, 1235, 1327. A favorable report upon Marquardt's method is given by G. LUNGE, V. MEYER, and E. SCHULZE, 1884: Chem. Cent., p. 854; Jour. Chem. Soc., 48, 708. 320 GALLIC ACID. the distillate with barium carbonate, and redistillation. To pu- rify ordinary chloroform it is necessary to repeat the process several times. In the qualitative use of the test 30 to 40 grams of the spirit are diluted with water so as to contain 12 to 15 per cent, of al- cohol, and the liquid shaken up with 15 c.c. of pure chloroform. The chloroform solution is washed with about an equal volume of water, and, after the latter has separated, is evaporated to dry- ness. To the residue are added a little water, one or two drops of sulphuric acid, and permanganate of potassium enough to give a red color after 24 hours. The estimation of fusel oil by measurement of the increase of volume of the chloroform layer, after shaking out, making comparison with a standard spirit, has been undertaken by B. ROSE, and advanced by Messrs. STUTZER and REITMAIR,' but the method is not well sustained. 8 The quantities of fusel oils present in alcoholic liquors have not been generally obtained upon trustworthy data. In certain whiskeys there were obtained, in the analyses of DR. DupRE,from 0.18 to 24 parts of amyl alcohol to 100 parts of ethyl alcohol. IS". P. HAMBERG 3 has given figures for the fusel oil in beer as follows : 1.14 gram of fusel oil in 100 liters of the beer, or about 0.00114 per cent. GALLIC ACID.-C 7 H 6 5 = C 6 H 2 (CO 2 H)(OH) 3 = 170. A trihydroxy-benzoic acid [CO 2 H : OH : OH : OH = 1 : 3 : 4 : 5]. Gallussaure. Found in nutgalls, sumach, tea, and, accompany- ing tannins, in a large number of plants. Gallic anhydride, as digallic acid, occurs in gallotannic acid, and gallic acid is a natu- ral fermentation product of certain glucoside-tannins. It is in use in medicine, as a reducing agent in photography, and in hair dyes. It is not a tanning agent. Gallotannic acid, in the human body, is soon converted into gallic acid. The gallic acid of commerce is wholly manufactured from the tannin of galls. Gallic acid, as a separate solid, is identified by its sensible properties and decomposition products (a); in solution, by its 1 Summary by UFFELMANN, 1886: Ding. pol. Jour., 261, 439; Jour. Chem. Soc., SO, 1079. B. ROSE, 1885: Archiv d. Phar., [3], 23, 62. * LUNGE, MEYER, and SCHULZE, 1884: Jour. Chem. Soc., 48, 709. 3 1885: Trans. Royal Acad. Stockholm; Schmidt's Jahrbucher der Medicin, 201, 27. GALLIC ACID. 321 reactions with iron salts, lime, and antimony, and its reducing power (d). It is distinguished from tannins by non-precipitation of gelatin, alkaloids, and antimony in presence of ammonium chloride. Separations from tannins, various acids, and from metals are indicated in e, methods of estimation in f, and tests for purity in g. a, 5. Gallic acid crystallizes, as C 7 H 6 O 5 .H 2 O = 188 (9.5# crystal water), in gray- white silky needles, or triclinic prisms, odorless, of an astringent and slightly acidulous taste, and acid reaction. The crystals are permanent, but lose all their water at 100 C. If the dry acid be gradually heated, in a glass tube, at 210' to 215 C. (4rlO-119 F.), a white or yellowish- white subli- mate of pyrogallol appears, in droplets, crystallizing on cooling : C 7 II 6 O 5 = C 6 H 6 O 3 + CO 2 . A dark residue remains. Heated quickly, in a porcelain capsule, at about 250 C. (482 F.), meta- gallic acid, C 6 H 4 O 2 , is formed, iri a black lustrous residue, sol- uble in strong alkali solution, with dark-brown color. Heated very gradually, with concentrated sulphuric acid, in a test-tube, to about 150 C. (302 F.), the mass turns wine-red to carrnine- red. If now cooled and poured into water, the latter will be colored yellow-brown, and a red-brown precipitate of ruh'irallol (rufigallic acid) partly crystalline (in rhombohedrons) will ap- pear. If the precipitate be washed and dried, then treated with very strong potassium hydroxide solution, a blue color changing to violet is obtained. Traces of rutigallol may be taken up with acetic ether. Baryta water also gives a blue color with rufigal- lol. If gallic acid be warmed with potassium hydroxide, tanno- melanic acid, of a black color, is produced. All alkaline solu- tions of gallic acid soon darken in the air. c. Gallic acid is soluble in 100 parts of cold or 3 parts of boiling water, the hot-saturated solution giving abundant crys- tals on cooling. It is freely soluble in alcohol, soluble in 39 parts of absolute ether, freely soluble in acetic ether, scarcely at all soluble in chloroform, benzene, or benzin. Gallic acid, by structure monobasic, is stated to form three classes of metallic salts by the displacing of one, two, and three atoms of its hy- drogen, only the alkali salts being soluble in water. Aqueous solutions of the acid soon decompose, with deposition of humus- like products. d. Solution of lime, added to an alkaline reaction, causes a white turbidity, changing to blue, later to green. Acetate of lead gives an abundant white precipitate, not especially char- acteristic. Ferric salts, and, more perfectly, the ferroso-ferric 322 GALLIC ACID. solutions, with free gallic acid, give a deep blue or blue-black pre- cipitate, decolored by boiling (with reduction of ferric to ferrous salt), and decolored by sufficient acetic acid or by excess of al- kalies. Tartrate of antimony and potassium causes a pre- cipitate, which, in distinction from gallotannin, is prevented or dissolved by ammonium chloride. Precipitates are not ob- tained with gelatin, or albumen, or starch, or with the alkaloids (all distinctions from gallotannin). Lead acetate with free gallic acid gives a bulky white pre- cipitate, which by warming condenses to a heavy powder, easily washed. The fresh precipitate, with sodium or potassium hy- droxide, turns red. As a reducing agent gallic acid is in general only a very lit- tle less forcible than tannin. Permanganate is promptly de- colored by gallic acid. Fehling's solution is turned from blue to yellow-brown at once, but the cuprous precipitate is very slowly obtained after heating for some minutes. Silver nitrate is slowly reduced, after warming, sometimes in part as a mirror. Molyb- date of sodium or ammonium reacts as with gallotannin. Ferric salts are partly reduced by boiling with gallic acid. e. Gallic acid may be separated from tannin by full precipi- tation of .the latter with cinchonine sulphate and filtration. By precipitation with some excess of gelatin, filtering (concentrat- ing the filtrate if need be), adding alcohol enough to throw down all the gelatin, and filtering again. Also by digestion with rasped hide. From the fruit acids, with tannin, if it be present, by calcium acetate, acetic acid and alcohol, or by acetic ether so- lution. From all acids not precipitated with lead acetate by this reagent, as above, separating the lead from the precipitate by treatment with hydrosulphuric acid and filtration. Recent lead hydrate removes all but a trace of free gallic acid. From its iron salts it is obtained by treatment with oxalic acid to fully change the color, and extraction of the mixture with acetic ether. f. Quantitative. The estimation of gallic acid, in solution free from other oxidizing Agents, may be accurately done by ti- tration with permanganate, in presence of indigo solution, as in Lowenthal's method for tannin (see under Tannin). 1 The gallic acid consumes more permanganate than an equal weight of tan- nin does, and more than does the quantity of tannin from which -it could be obtained. 3 The permanganate solution may be stand- ardized by freshly dissolved weighed gallic acid of given purity. 1 LOWENTHAL, 1877: Zeitsch. an. Chem., 16, 39. 2 PROCTER, Chem. News, 36, 60. GLYCERIN. 323 Tannin, if present, must be first removed as in Luwenthal's method. Gallic acid may be estimated, in absence of tannin, by the increase of weight of zinc oxide. A weighed quantity of the oxide, freshly ignited, is digested with the solution of free gallic acid, filtered out, washed, dried at 110-120C., and its increase of weight taken as gallic acid. A method, after FLECK, by pre- cipitation as copper salt, giving approximate results in presence of tannin, is conducted as follows : The prepared solution is fully precipitated with a filtered solution of cupric acetate ; the pre- cipitate washed and then exhausted with cold solution of car- bonate of ammonium. The last solution, containing all the gallate of copper with a very little tannate, is evaporated to dryness, the residue moistened with nitric acid, ignited, and weighed as oxide of copper. This weight multiplied by 0.9 gives the quantity of gallic acid, the full ratio being 0.9126, but allowance is made for solution of a little tannate by the carbonate of ammonium. The ratio between oxide of copper and tannic acid is 1.304. g. "An aqueous solution of gallic acid should not precipi- tate alkaloids, gelatin, albumen, gelatinized starch, or solution of tartrate of antimony and potassium with chloride of ammonium (distinction from tannic acid) " (U. S. Ph.) GALLOTANNIN. See TANNINS. GLYCERIN. Glycerol. C 3 H 8 O 3 = 92. (C 3 H 5 )'"(OH) 3 . Propenyl, C 3 H 5 =: CH 2 . CH . CH 2 , is a residue of propane, the third member of the marsh-gas series. Glycerin is produced, along with candle-manufacture and the production of the fat acids ( " stearin " and " olein " ), by saponification of the fats, with water as superheated steam, or with lime, or with sulphuric acid. It occurs also among the products of the alcoholic fermentation of sugar. Glycerin taken alone is recognized by its sensible properties (a) ; in dilute forms or in certain admixtures it is revealed by the bead-test with borax, its power of neutralizing boracic acid, arid the odor of its vapors when strongly heated (d). As a re- ducing agent it affects permanganate promptly in alkaline mix- ture, scarcely at all in neutral or acid liquids, and does not alter Fehling's solution. It is separated from substances more vola- tile than water by their distillation, and from non- volatile sub- stances by its own distillation at a heat a little above that of the water- bath ; from matters insoluble in alcohol, by use of this sol- 324 GLYCERIN. vent, best by use of lime and alcohol (e). It is estimated gravi- metrically by careful evaporation with alcohol -ether ; volumetri- cally by the permanganate reaction, forming oxalic acid (/'). Tests and authorized standards of purity (g, p. 328). a. Glycerin is a colorless, clear, syrupy liquid, capable of crystallization in low winter temperatures, taking forms of the rhombic system, or congealing in white, crystalline masses, nearly or quite anhydrous, the melting point being 22 C. Specific gravity at 15 C., taking water at same temperature as standard, 1.26468 (MENDELEJEFF), 1.2653 (GERLACH); at 15 C., taking water at 0C., 1.26358 ; at 17.5 C., 1.262 (STROHMER). Glycerin is very hygroscopic, and at ordinary temperatures it vaporizes in only the slightest degree, but at 100 C. it vaporizes or distils to .a sensible extent. At this temperature and 760 millimeters barometric pressure it has a vapor tension of 64 millimeters. At 290 it boils with partial decomposition, evolving vapor of acro- lein, C 3 H 4 O. With superheated steam at 180 to 200 C. it dis- tils completely. Evaporated in an open dish at 150 to 200 C., when perfectly pure, it leaves no residue behind. Heated in a capsule, at 92 C. vapor rises almost imperceptibly, at 100 C. quite perceptibly, at 130 C. abundantly, without irritating pro- ducts to a sensible extent at last-named temperature (TRIMBLE, 1885). b. Glycerin has a pure sweet taste of much intensity, with- out odor. Undiluted it has a heating effect when applied to the surface. c. Exposed to the air glycerin absorbs water, finally to the extent of about 50 per cent., and is soluble in all proportions of water and of alcohol. In mixture with water the volume is re- duced and the temperature raised, the greatest liberation of heat being obtained with 58 parts of glycerin to 42 parts of water, in which proportions the contraction of volume is about 1.1 per cent., and the elevation of temperature about 5 C. In ether, glycerin is slightly soluble, 1 part of glycerin of sp. gr. 1.23 re- quiring 500 parts of ordinary ether for solution. It is soluble in a mixture of 3 parts of alcohol and 1 part of ether ; also in a mixture of 2 volumes of absolute alcohol and 1 volume of com- mon ether. Not soluble in chloroform, benzene, or fixed oils. Soluble in a mixture of equal weights of chloroform and alcohol. Glycerin when pure is neutral to all indicators. Glycerin dis- solves the alkaline earths to a considerable extent, with chemical combination. If the solutions be charged with carbon dioxide the earths are mainly precipitated. With lead it forms the GL YCERIN. 325 flyceride, C 3 H 6 PbO 3 , crystallizable in fine white needles. So- ium glyceride, C 3 H 7 NaO 3 , is a white, hygroscopic powder, resolved by water into glycerin and sodium hydroxide. d. " If a fused bead of borax on a loop of platinum wire be moistened with glycerin previously made slightly alkaline with diluted solution of soda, and after a few minutes held in a color- less flame, the latter is tinted deep green." Glycerin abstracts boric acid from borax, so as to affect the reaction to litmus. If a borax solution be colored (blue) with litmus, and a solution con- taining glycerin, neutral in reaction, be also colored with blue litmus, on mixing the solutions a red color will be obtained. Warming restores the blue color, but the red reappears when the liquid is cool again. If a portion of glycerin be heated to boil- ing in a dry test-tube, the characteristic acrid vapors of acrolein will be obtained. If in aqueous mixture, the glycerin must be concentrated for the test, which is also rendered more delicate by the addition of a little dry phosphoric acid or potassium bi- sulphate. Glycerin alone is not carbonized by heating with either of these agents. If 2 drops of glycerin (free from bodies carbonized by sulphuric acid) be mixed with 2 drops each of melted phenol and sulphuric acid, and heated somewhat over 120 C. to the production of a resinous mass, and when cold am- monia be added, a fine carmine-red color will be obtained. Permanganate solution, acidulated with sulphuric acid, is but very slowly decolored by glycerin, and even by boiling heat the oxidation of the glycerin is difficult. But in alkaline solu- tion of the permanganate decoloration by glycerin is prompt, the reaction being as follows (BENEDIKT and ZSIGMONDY) : C 3 H ? O 3 + 2K 2 Mn 2 O 8 K 2 C 2 O 4 + K 2 CO 3 + 4MnO 2 + 4II 2 O. Fehling's solution is but very slightly reduced by glycerin. A quite concentrated solution of pure glycerin, boiled 10 minutes with Fehling's solution, and then set aside for 24 to 48 hours, yields some precipitate of the cuprous hydroxide. But dilution of the glycerin with ten volumes of water prevents the reaction. e. Separations. Glycerin in watery mixture is not concen- trated by evaporation without loss, neither is any part of it ob- tained anhydrous by continued evaporation. From 100 to 130 C. glycerin alone distils unchanged (TKIMBLE, 1885). From so- lution in alcohol it can be recovered almost without loss in the residue left by gentle evaporation with additions of a little ab- solute alcohol from time to time. Glycerin may be separated from fluid extracts of vegetable drugs, also from wines, or from sugars and gums, as follows : Slaked lime is added, the liquid is 326 GLYCERIN. treated with a little alcohol and evaporated to dryness at a very gentle heat, adding small portions of alcohol in completing the evaporation, and then exhausted with several portions of alcohol of about ninety per cent, strength by weight. Tiie clear alcoho- lic solution is evaporated, with the alcoholic filter-washings if filtration has been required, and the residue will be approxi- mately pure glycerin. In the case of certain extracts, and as a precaution in separating from unknown matters, it is desirable to take up the glycerin residue with a mixture of two volumes of absolute alcohol and one volume of stronger ether, filtering if need be, and evaporating again. With quantitative precautions in completing the extractions, and washing filters, then evapo- rating in weighed beakers, the operation will serve as a good practical estimation, though of course absolute glycerin is not obtained for weight. Instead of the ether-alcohol, a mixture of equal weights of chloroform and alcohol does well in some cases as a solvent for purification, but the former is generally the best. The lime is an essential aid in holding sugars, gums, and many extractive matters, from solution in the ninety-per-cent. alcohol. 1 From soaps glycerin is obtained by acidulating with dilute sulphuric acid for removal of the fat acids, addition of ba- rium carbonate for removal of excess of sulphuric acid, when al- cohol is added, alkali sulphates filtered out, and evaporation con- ducted with the addition of alcohol and filtration. f. Quantitative. Estimation of glycerin of fats. About 30 grams of the dried and filtered clear fat, with 15 grams potas- sium hydroxide or 21 grams sodium hydroxide dissolved in the least sufficient amount of water, and 50 c.c. alcohol, are digested ir rhe use of the lime with the alcohol was reported upon, for estimation of the glycerin of wines, by E. REICHARDT, 1875: Archiv d. Phar., [3], 7,408: Zeitsch. anal. CJiem. (1878), 17, 109. An elaborate examination of this method of separating glycerin from wines was made by NEUBAUER and BORGMAN, 1878: Zeitsch. anal. Chem., 17-, 445. These authors found a considerable impurity in the glycerin extracted from wines by Reichardt's directions, but they were able to recover very nearly the quantity of glycerin added to wines. Control analyses by Reichardt's method, in separation from cane sugar, grape sugar, and medicinal fluid extracts, were made by the author and another (A. B. Prescott and 0. H. Koehnle) in 1878: New. Bern., New York, 7, 354. From mixture with sucrose 99.8 per cent, of the glycerin taken was recovered : from mixture with glucose 99.7 per cent, was recovered : in both cases the glycerin being obtained free from sugar. From fluid extracts of cinchona and of gen- tian, to which sugars had been added, there were separated 97.6, 98.6, and 95.4 per cent, of the glycerin added. The ether-alcohol mixture mentioned in the text separated pure glycerin from sucrose, but not from glucose, some of which was taken up with the glycerin. The same was found to be true of the chloroform-alcohol mixture referred to in the text. GLYCERIN. 327 over the water-bath to perfect saponification. The alcohol is evaporated off, the soap dissolved in water, diluted sulphuric acid is added, and the mixture warmed, until the separation of the surface-layer of fat acids is complete. When cold the fat crust is removed, the aqueous liquid filtered, and the fat acids washed on the filter ; or the fat acids are filtered out as in Heh- ner's method (p. 250). The excess of sulphuric acid in the fil- trate is taken up with barium carbonate, the nitrate evaporated, with additions of alcohol, to a thin syrupy consistence, then treated with a mixture of 3 parts of 95$ alcohol and 1 part of ether and filtered, washing the filter with the same mixed sol- vent. This filtrate is evaporated in a platinum dish, at 100 C., until two weighings do not differ over 3 to 5 milligrams, or dried in a desiccator to a constant weight. The residue is then ignited, and the weight of the ash deducted. The glycerin of strictly neutral fats can be calculated from Kottstorl'er's number the milligrams of potassium hydroxide to saponify 1 gram of fat. Taking these milligrams as thousandths of a gram, 3KOH : C 3 H 8 O 3 : : 168 : 92. Estimation by oxidation with p&rmanganate (BENEDIKT and ZsiGMONDY. 1 ) The reaction is stated under d, and the resulting oxalic acid is the measure obtained. The fat is saponified with potash and methyl alcohol, the alcohol evaporated off, the residue dissolved by hot water, the soap decomposed by diluted hydro- chloric acid, and the fat acids melted and set aside until fully clear. Some hard paraffin is added to the fat acids, the mixture cooled and filtered, and the residue washed. The filtrate is neu- tralized with potassa, and is now ready for the reaction. - First add 10 grams potassium hydroxide, 2 and then add, at ordinary temperature, of a permanganate solution of about 5$ strength sufficient to render the liquid no longer green but blue or black in color. The mixture is heated to "boiling, with precipitation of manganese dioxide, and then enough sulphurous acid is added to make the red liquid colorless, and the mixture filtered through a wet filter large enough to contain at least half at once, and the residue well washed with boiling water. If the last washings be turbid with manganese dioxide, a little acetic acid is added. The liquid is now heated to boiling, and fully precipitated by calcium acetate or chloride. The oxalate of calcium precipitated is esti- : Analyst, 10, 205, from Chem. Zeit. An investigation of this and other methods, now being made by A. J. BAUMHARDT and the author, will be communicated at an early date. 2 Fox and WANKLYN (1886 : Chem. News, 53, 15) take a quantity of ma- terial containing not over 0.25 gram of glycerin, and add 5 grams of KOH. 328 GLYCERIN. mated at the discretion of the operator. But inasmuch as the precipitate is liable to contain, as impurities, calcium silicate and calcium sulphate, it is better to estimate the oxalate volurnetri- cally with permanganate, or, after ignition, by titration with half normal hydrochloric acid, using diraethylanilin orange as an in- dicator. The hydrochloric acid may be standardized by anhy- drous sodium carbonate. 106 parts of Na 2 CO 3 , or 72.8 parts of HC1, indicate 90 parts of H 2 C 2 O 4 , or 92 parts of glycerin. In this operation methyl alcohol is used because ethyl alcohol, if employed, is not wholly expelled, and suffers oxidation by per- manganate in alkaline liquid with formation of oxalic acid. So- luble fat acids do not interfere. The method is applicable to any ordinary neutral mixture of glycerin. Benedikt and Zsig- mondy obtained the following percentages of glycerin: From olive oil, 10.15-10.38; linseed oil, 9.45-9.97 ; tallow, 9.94-9.98- 10.21; butter, 11.59; Japan wax, 10.3-11.2; beeswax, 0. g. Tests of purity. The U. S. Ph. prescribes as follows: " Glycerin should be neutral to litmus-paper. Upon warming a portion of 5 or 6 grams with half its weight of diluted sulphu- ric acid, no butyric or other acidulous odor should be obtained. A portion of 2 or 3 grams, gently warmed with an equal volume of sulphuric acid in a test-tube, should not become dark- colored. A portion of about 2 grams, heated in a small open porcelain or platinum capsule, upon a sand bath, until it boils, and then ignited, should burn and vaporize so as to leave not more than a dark stain (absence of sugars and dextrin, which leave a porous coal). A portion heated to about 85 C. (185 F.) with test solution of potassio-cupric tartrate should not give a decided yellowish-brown precipitate, and the same result should be obtained if, before applying this test, another portion be boiled with a little diluted hydrochloric acid for half an hour (absence of sugars). After full combustion no residue should be left (metallic salts). Diluted with ten times its volume of distil- led water, portions should give no precipitates or colors when treated with test solutions of nitrate of silver, chloride of barium, sulphide of ammonium, or oxalate of ammonium (acrylic, hy- drochloric, sulphuric, and oxalic acids, iron and calcium salts)." " Shaken with an equal volume of sulphuric acid, no colora- tion, or only a very slight straw coloration, should result. When gently heated with diluted .sulphuric acid, no rancid odor is produced. Sp. gr. about 1.25 " (Br. Ph.) "Sp. gr. 1.225 to 1.235. Heated in an open dish to boil- ing, and then ignited, it should burn without residue. It should HYDRASTINE. 329 not reduce ammoniacal solution of silver nitrate, at ordinary tem- perature, within half an hour. Warmed with an equal volume of sodium hydrate solution (15$), it should not be colored, nor should ammonia be developed ; and gently warmed with diluted sulphuric acid, no disagreeable rancid odor should be given " (Ph. Germ.) Sp. gr. 1.242. Undergoes complete combustion, leaving no residue (Ph. Fran.) Impurities : lead salt, lime, lime sulphate, sodium chloride, oxalic acid, butyric acid. Adulterations : ex- cess of water, dextrin, glucose syrup, honey (Ph. Fran.) The sulphuric acid test is doubtless severe enough if the directions of the U. S. Ph. be followed with omission of the gentle warming, as sufficient elevation of temperature results from the admixture. The silver nitrate test is much influenced by the conditions of time and light. If treatment with test solution of silver nitrate for half an hour, in the dark, be adopt- ed, the test is certainly not too severe. The combustion test is efficient for the exclusion of carbohydrates. Messrs. Patch, Warder, and Goebel have each lately reported upon the quality of glycerin sold in the United States. 1 GUARANINE. See CAFFEINE, p. 77. HOMATROPINE. See MIDKIATIC ALKALOIDS. HOMOQUININE. See CINCHONA ALKALOIDS, p. 92. HYDRASTINE. C 22 H 23 NO 6 = 397 (MAHLA, 1863). The colorless alkaloid of Hydrastis Canadensis, or " golden seal " root, in which it accompanies the yellow alkaloid berberine. Hydras- tine is also a commercial name for medicinal preparations of the yellow alkaloid berberine, from Hydrastis. 2 Perrins obtained 1 Proc. Am. Pharm. for 1885: 33, pp. 481, 484, 485. 2 The colorless alkaloid of Hydrastis was announced as a crystallizable al- kaloidal body, in 1851, by DURAND, of Philadelphia, who proposed the name "hydrastine," but was left in doubt because unable to obtain crystallizable salts. The name "hydrastine " had been given to the yellow alkaloid of golden seal by RAFINESQUE in his "Medical Flora of the United States," vol. i., 1828. In Europe the yellow alkaloid had been found in other plants and named " jamaicine "*by HUTTENSCHMID in 1824, " xanthopicrite " by CHE- VALLIER and PELLETAN 'in 1826, and ' berberine " by BUCHNER and HERBERGER in 1830, with a better description by FLEITMAN in 1847. The yellow alkaloid was found in Hydrastis by DURAND in 1851, and identified with the yellow al- kaloid of Berberis and other plants by MAHLA as late as 1862, and by'PERRiNS, 1863. Prof. LLOYD states (1884) that " there is little indication that the term hydrastine," as applied to the yellow alkaloid of Hydrastis, " will be sup- 330 HYDRASTINE. \\ per cent, from the dried root. Lloyd states the yield in manu- facture to be J to } per cent. Hydrastine is characterized by its crystalline form when free, and the amorphous condition of its salts (&), with the reactions it gives as an alkaloid (d). It is prepared from golden seal as di- rected (tf), and estimated gravimetrically (/"). a. Hydrastine forms four-sided prisms, orthorhombic, lus- trous when perfect, usually broken and opaque white. The crys- tals melt at 135 C. (MAHLA), at 1 32 C. (POWEK), and in strong heat decompose with odor of phenol. The salts of hydras- tine refuse to crystallize. The hydrochloride is anhydrous, C 22 H 23 ]SrO 6 .HCl; also the sulphate, (C 22 H 23 NO 6 ) 2 H 2 SO 4 .- Hydrastine is levo-rotatory, with a specific rotatory power, in chloroformic solution, of [a] D= 170 (POWER). b. Hydrastine, free, is tasteless and odorless. The salts have an acrid taste. Hydrastine is the true active principle of Hydrastis Canadensis (Prof. BARTHOLOW, 1885 '). Three grains of the hydrochlorate caused the death of a frog in four minutes, and the results upon rabbits were corresponding. Like strych- nine, it causes death by arrest of the respiratory movements in a tonic spasm. c. Hydrastine is not appreciably soluble in water or in di- lute alkaline solutions. At 15 C. it dissolves in 1.75 parts of chloroform, in 15.7 parts of benzene, in 83,46 parts of ether, planted by berberine at any immediate day": "Drugs and Medicines of North America," vol. i. 100. A. B. DURAND, 1851: Am. Jour. Phar., 23, 112. W. S. MERRELL, 1862: Am. Jour. Phar., 34, July. J. D. PERKINS, 1802: Phar. Jour. Trans., [2], 3, 546 (May): first separation as a colorless alkaloid. F. MAHLA, 1886: Am. Jour. Sci., [2], 36, 27. J. U. LLOYD, 1878: Proc. Am. Pharm., 26, 805. F. B. POWER, 1884: Proc. Am. Pharm., 32, 448. J. U.LLOYD, a full history: "Drugs and Medicines of North America," 1884, vol. i. 130. FREUND and WILL, a critical investigation, 1886: Ber. d. chem. (res., 19, 2797; Phar. Jour. Trans., 16. Upon a second colorless alkaloid in Hydrastis see A. K. HALE, Ann Ar- bor, 1873: Am. Jour. Phar., 45, 247 J. C. BURT, 1875: Am. Jour. Phar , 47, 481. H. LERCHEN, Philadelphia, 1878: Am. Jour. Phar., 50, 470. J. U. LLOYD, 1884: " Drugs and Med. of North America," vol. i. 139. According to FREUND and WILL (1886, loc. cit.) hydrastine has decided chemical resem- blance to narcotine (CasHssNOv). When oxidized by permanganate or by di- lute nitric acid, hydrastine was found to yield a crystalline acid identical with opianic acid, together with a crystalline base closely resembling cotarnine (compare under Narcotine, d). 1 Communication, from physiological trials, in "Drugs and Medicines of North America," i, 156. HYDRASTINE. 331 and in 120 parts of alcohol ( POWER, 1884 '). It does not dissolve in petroleum benzin. The ordinary salts are soluble in water ; the tannate and picrate insoluble. The salts of hydrastine mostly have an acid reaction. The acetate, formed in solution, decom- poses on evaporation. d. Alkali hydrates precipitate hydrastine, from the aque- ous solution of its salts, in a bulky amorphous mass, which finally takes on crystallization, with great reduction of volume. The precipitate is but slightly soluble in excess of alkalies. White precipitates are produced by potassium iodide, potassium fer- rocyanide, sulphocyanide, mercuric chloride, and by tannic acid (POWER). The general reagents for alkaloids cause preci- pitates iodine in potassium iodide, brown ; Mayer's solution, white ; platinic chloride, orange-yellow ; gold chloride, yellowish- red ; picric acid, yellow. Potassium bichromate gives a yel- low precipitate. Sulphuric acid, undiluted, causes a yellow color, becoming red on warming, and turning to brown on adding a crystal of bichromate. Concentrated sulphuric acid and molybdate of ammonium give an olive-green color (POWER). Concentrated nitric acid produces only a yellowish color in the cold. The hydrochloride solution, treated with chlorine, shows blue fluorescence (MAHLA). Ethyl-hydrastine was obtained in hydriodide by PowER, 2 who also formed a hydro-hydrastine C 22 Ho 7 NO 6 , in the hydrochloride. e. Hydrastine may be separated from golden seal root as fol- lows (POWER, LLOYD : 1884) : The powdered root is moistened with alcohol and percolated with the same solvent ; sulphuric acid in strong excess is added to the percolate ; after four hours the crystals of berberine sulphate are filtered out ; ammonia is added to the filtrate until it has but a slightly acid reaction and the crystallized ammonium sulphate is filtered out ; the filtrate is concentrated (by distillation) to a syrupy consistence, and the residue poured into ten times its volume of cold water. After twenty four hours the precipitated resinous substances, oils, etc., are filtered out ; ammonia- water in decided excess is added to the filtrate ; and the resulting precipitate, impure hydrastine, collected and dried. The product is digested with 100 times its weight of cold water, to which sulphuric acid has been carefully added to slight acid reaction ; after twenty-four hours the liquid is filtered and ammonia in excess added to the filtrate, the pre- cipitate collected on a strainer, dried, and then powdered and 1 Proc. Am. Pharm., 32, 450. 2 1884: Proc. Am. Pharm., 32, 454. 332 HYDRASTINE. extracted with boiling alcohol. On cooling the solution gives crystals of hydrastine, still dark yellow with impurities, and to be recrystallized from alcohol, repeatedly, until perfectly color- less. f m Quantitative The free alkaloid crystallizes anhydrous, C 22 H 23 NO 6 , and the crystals or the well- washed precipitate by ammonia, when obtained colorless, may be dried at 100 C. for weight. The gold chloride of hydrastine, by precipitation of the hydrochloride of the alkaloid with auric chloride, and drying at 100 C., gave Prof. Power J 16.92 per cent, of metallic gold. The formula, (C 22 H 23 \N~O 6 . HCl) 2 Au01 3 = 1 1 69.2, indicates 16.78 per cent, of gold and 67.91 per cent, of hydrastine. The platinum chloride, obtained by precipitation of the hy- drochloride solution, gave MAHLA 16.17$ of platinum ; the for- mula (C 22 [-l 23 NO 6 .HCl) 2 PtCl 4 indicating 16.120 of platinum. HYDROQUININE. See CINCHONA ALKALOIDS, p. 91. HYGRINE. See COCA ALKALOIDS, p. 173. HYOSCYAMINE. See MIDRIATIC ALKALOIDS. IGASURINE. See STRYCHNOS ALKALOIDS. INKS. See TANNINS. JAPACONITINE. See ACONITE ALKALOIDS, p. 18. KATRINES. See CINCHONA ALKALOIDS, p. 166. LANTHOPINE. See OPIUM ALKALOIDS. LARD. See FATS and OILS, p. 290. LINOLEIC ACID. See p. 249. LINSEED OIL. See p. 284. MADDER RED. See COLORING MATTERS, p. 189. MAGENTA. See p. 191. 1 1884: Proc. Am. Pharm., 32, 453. MALIC ACID. 333 MALIC ACID. H 8 C 4 H 4 5 = 110. C.O 2 H . CH 2 . CHOH . OO 2 H. Aepfelsaure. Distributed widely through the vegetable kingdom. Reported already in not less than 200 plants (Huse- mann's u Pflanzenstoffe "). Most abundant in fruits, but found in other parts of plants. 1 Usually obtained from mountain-ash berries or from unripe apples. LENSSEN (1870) obtained 6.62$ from barberry berries, only 1.58$ from mountain-ash berries. Others report about 2$ from the latter. It is abundant in to- bacco. It is believed identical with Minispermic acid, Solanic acid, Tannacetic acid, Euphorbic acid, and perhaps with Igasuric acid. It is formed artificially from asparagin, from tartaric acid, and from succinic acid. It is not manufactured for use. Malic acid is identified, more especially, by its sublimation products (0), the deportment of its lead precipitate when warmed and when treated with ammonia, and the formation of its cal- cium precipitate by alcohol, also by its reduction of dichromate with apple-odor (d). It is separated from citric, tartaric, and oxalic acids by non -precipitation in boiling calcic aqueous solu- tions, or by the alcohol solubility of its ammonium salt (e) ; from fruit juices by systematic treatment (e). Estimated, gravimet- rically, as lead salt (/"), or as calcium salt weighed as sulphate. Methods of preparation are indicated at g. a. Malic acid crystallizes with some difficulty, and from syrupy solution, in tufted needles or in four or six sided prisms, anhydrous, and deliquescent in the air. Heated in a small retort over an oil-bath or sand-bath to 175 or 180 C., malic acid evolves vapors of maleic and fumaric acids, which crystallize in the retort and receiver. The fumaric acid forms slowly at 150 C., and mostly crystallizes in the retort in broad, colorless rhombic or hexagonal prisms, which vaporize without melting at about 200 C., to condense in needles, and are soluble in 250 parts of water, easily soluble in alcohol or ether. If the temperature is suddenly raised to 200 C. the maleic acid is the chief product. Maleic acid crystallizes in oblique, rhomboidal prisms, which melt at 130 C. and vaporize at about 160 C., condensing in hard needles, and are readily soluble in water, alcohol, and ether. The test for malic acid, by heating to 175 or 180 C., may be made in a test-tube, with a sand-bath, the sublimate of fumaric and maleic acids condensing in the upper part of the tube. Malic acid melts below 100 C., and does not lose weight at 120 C. ; at 1 For list of plants containing malic acid see Grinelin-Kraut's " Handbuch," v. 336, Supplem. 884. 334 MALIC ACID. the temperature of the test water-vapor is separated maleic and f umaric acids both having the ultimate composition of malic an- hydride (C 4 H 4 O 4 ). Solution of malic acid quickly moulds, with various products. Fermented with yeast or cheese, in presence of calcium carbonate, succinic acid is formed, with acids of the formic series. b, c. Malic acid is freely soluble in water, alcohol, and ether ; the malates soluble or sparingly soluble in water, mostly insolu- ble in alcohol ; ammonium malate soluble in alcohol. Malic acid is dibasic, and forms normal and acid salts. Like tartaric and citric acids, it prevents the precipitation of metals by alkalies. d. Solution of acetate of lead precipitates malic acid, more perfectly after neutralizing with ammonia, as a white and fre- quently crystalline precipitate, which upon a little boiling melts to a transparent, waxy semi-liquid (a characteristic reaction, ob- scured by presence of other salts). The precipitate is very sparingly soluble in cold water, somewhat soluble in hot water (distinction from Citrate, Tartrate, and Oxalate), crystallizing out when cold ; soluble in strong ammonia, but not readily dis- solved in slight excess of ammonia (distinction from citrate and tartrate) ; slightly soluble in acetic acid and in malic acid. If the precipitate of malate of lead is treated with excess of am- monia, dried on the water-bath, triturated and moistened with alcoholic ammonia, and then treated with absolute alcohol, only the malate of ammonium dissolves (distinction from Tartaric, Citric, Oxalic, and many other organic acids, the normal ammo- nium salts of which are insoluble in absolute alcohol). Also, malic acid maybe separated from tartaric, oxalic, and citric acids, in solution, by adding ammonia in slight excess, and then 8 or 9 volumes of alcohol, which leaves only the malate of ammonium in solution. Solution of chloride of calcium does not precipitate malic acid or malates in the cold (distinction from Oxalic and Tartaric acids) ; only in neutral and very concentrated solutions is a pre- cipitate formed on boiling (while calcic citrate is precipitated in neutral boiling solutions, if not very dilute). The addition of alcohol, after chloride of calcium and boiling, in neutral solu- tion, produces a white, bulky precipitate of calcic malate in even dilute neutral solutions (indicative in absence of sulphuric and other acids whose calcium salts are less soluble in alcohol than in water). The precipitate dissolves in water, and is reproduced by alcohol. Solution of mercurous nitrate gives a white, flocculent pre- MALIC ACID. 335 cipitate, slightly soluble in water (formed in solutions not very dilute), not soluble in malic acid, but dissolving in dilute acetic acid, in sodium malate solution, and in excess of the precipitant. Silver malate precipitate darkens but slightly on boiling. Permanganate of potassium is reduced but very slowly (distinc- tion from Tartaric acid) ; somewhat more on addition of sulphuric acid. Nitric acid, on boiling, is readily deoxidized by malic acid, brown vapor appearing. Dichromate of potassium solution is reduced, even in the cold. By addition of dilute sulphuric acid, and warming, apple odor is developed, according to PAPASOGLI and PoLi, 1 who distinguish between malic, citric, and succinic acids, with use of this reaction, as follows : The acid, if neces- sary, may be first precipitated with calcium chloride and alcohol ; the precipitate, freed from alcohol, treated with dilute sulphuric acid and the calcium sulphate precipitate filtered out. The fil- trate, containing sulphuric acid, is boiled with a little dichro- mate. If (1) the liquid remains perfectly yellow, succinic acid may be present ; (2) the color becomes yellowish-green, citric acid may be present, as well as succinic ; (3) if a green color ap- pears, with odor of ripe or over-ripe apples, malic acid is indi- cated. [A green color without apple odor would result from Tartaric acid and from numerous reducing agents which might be precipitated by calcium chloride with alcohol.] In the various oxidations of malic acid above mentioned, its products are formic acid, carbon dioxide and water, and, from nitric acid especially, oxalic acid. Dichromate in the cold concentrated solution pro- duces some Malonic acid (Dessaignes, 1858), C 3 H 4 O 4 , crystalliz- able, soluble in alcohol and ether. Malic acid is capable of reduction to succinic acid, by hydriodic acid, at 130 C., and by other agents. e. Malic acid can be separated from Citric, Tartaric, and Oxalic acids by the solubility of its ammonium salt in alcohol, as follows: 3 Ammonia is added to neutral reaction, the solution well concentrated, and again neutralized, treated with 7" or 8 volumes of 98$ alcohol, and set aside 12 to 24 hours, when it may be filtered, and the filtrate treated with lead acetate, for malate. The residue may contain ammonium citrate, tartrate, oxalate. The separation of the same four acids may be done, through 1 1877: Oazzettachim. ital, 7, 294; Jour. Chem. Soc., 32, 807; Jahr. Phar., 1878, 128. 2 BARFOED, 1868: Zeitsch. anal. Chemie, 7,408. In a full report upon the separations of malic acid, loc. cit., p. 403. 336 ; MALIC ACID. the calcium precipitates, with approximate closeness, by the fol- lowing method : l Solution of Oxalic, Tartaric, Citric, and Malic acids. (If sulphates are present, remove by just enough barium chloride with hydrochloric acid.) Add ammonium hy- drate to a slight alkaline reaction ; add ammonium chlo- ride solution ; then enough calcium chloride solution, and let stand from ten to twenty minutes. Filter. Precipitate (a) : Oxalate (complete), Tartrate (nearly complete). (If Phosphates are present, separate from oxalate by acetic acid, and identify by molybdate.) Filtrate (b) : Citrate, Malate. Wash Precipitate (#), digest it in the cold with sodium . hydrate solution (or potassium hydrate solution), then dilute a little and filter. Residue (c) : Oxalate. Nearly insoluble in acetic acid. Solution (d) : Tartrate. Boil some time. A precipitate indi- cates tartrate. Test by reducing power, with dichromate, silver salt, or permanganate, and by Fenton's color test. To Filtrate (b) which must have excess of calcium chloride add 3 times its measure of alcohol. If a pre- cipitate occurs, filter. Precipitate (e) : Citrate, Malate (nearly complete). Filtrate (/): (May contain benzoic, acetic, formic acids, etc.) Wash Precipitate (e) with alcohol ; dissolve on the filter with dilute hydrochloric acid. To the filtrate add ammo- nium hydrate to slight alkaline reaction, and boil for some time. If a precipitate occurs filter, hot (Filtrate A). Precipitate (g) : Citrate. Confirm by dissolving again with hy- drochloric acid, neutralizing with ammonia, and boiling, to obtain a precipitate. Other tests may be applied. Filtrate (A) : Malate. (May contain succinate.) Try for malic acid by precipitating with strong alcohol. Test a precipi- tate, so obtained, by reduction of dichromate, by lead pre- cipitate, and other tests. To separate from succinic add strong nitric acid and evaporate to dryness, when there will be oxalic acid instead of malic, the succinic acid un- changed. Test for oxalic by calcium salt. Malic acid may be separated from Tannic acid by digesting the solution a half -hour with well- washed rasped hide, and filter- ing out the tannate. The filtrate may be concentrated and treat- 1 Except final tests, arranged from Fresenius' "Qualitative Analysis," S. W. Johnson's edition of 1875, p. 304. MECONIC ACID. 337 ed with lead acetate, to be tested for malic acid. Or both acids may be precipitated by chloride of calcium, with a slight excess of ammonk and alcohol, and the inalate then washed out of the precipitate with water. 1 Also, tannic and gallic acid may be re- moved by acetic ether. For determining the presence of malic acid in Fruit Juices, the expressed juice or water extract is first precipitated by lead acetate solution, when the washed precipitate may be treated as directed under d, p. 334. f. The estimation of malic acid is usually done gravimetri- caliy, as a lead salt. The alcoholic solution of malate of ammo- nium may be precipitated with acetate of lead, washed with alcohol, dried on the water-bath, and weighed as malate of lead. PbC 4 H 4 O 5 : H 2 C 4 H 4 O 5 :: 1 : 0.3953. The crystals of this salt contain, three molecules of water of crystallization. The calcium normal malate precipitate, in strong alcohol, may be washed with alcohol, converted into sulphate, this washed with alcohol, dried, ignited, and weighed. CaSO 4 : H 2 C 4 H 4 O 5 . g. In the preparation of malic acid on the small scale, the lead precipitate may be decomposed by boiling with excess of very dilute sulphuric acid ; filtering, neutralizing one- half the filtrate with ammonia, and mixing this with the other half of the filtrate, then evaporating to crystallize, as ammonium acid malate, NH 4 HC 4 H 4 O 5 , in large orthorhombic prisms. 1 MARGARIC ACID. See FATS AND OILS, p. 244. MECONIC ACID. H3C 7 HO 7 == 200. Oxychelidonic acid. Found only in opium, which yields 3 to 4$ of it. Not manu- factured for use. Identified by its physical properties, its pro- ducts when heated, its precipitation by hydrochloric acid, and reactions with iron and other metals. It is separated from opium through formation of the calcium salt or lead salt. Meconic acid crystallizes in white, shining scales or small rhombic prisms, containing three molecules of crystallization water, fully given off at 100 C. At 120 C. (248 F.) dry meconic acid is resolved into comenic acid / at above 200 C. 1 Further for these separations, and for separation from Gallic, Benzoic, and other acids, see BARFOED where last cited. Separation from Gallic acid is directed by adding calcium chloride to the slightly alkaline solution, and leav- ing some time, without heat, for precipitation of calcium gallate. 2 For methods of preparation on larger scale, with first precipitation as lead salt, see " Watts's Dictionary," iii. 7*9; with first precipitation as calcium, salt, Husemann's "Pflanzenstoft'e," 537. 338 MECONIC ACID. the comenic acid is resolved into pyrocomenic acid and other products. The sublimate of comenic acid dissolves sparingly in hot water, not at all in absolute alcohol. It crystallizes in prisms, plates, or granules. Solution of comenic acid gives a red color with ferric chloride, green pyramidal crystals with cupric sul- phate in concentrated solution, and a yellowish white granular precipitate with acetate of lead. Meconic acid is soluble in 115 parts of water at ordinary temperatures, less soluble in water acidulated with hydrochloric acid, much more soluble in hot water, freely soluble in alcohol, slightly soluble in ether. It has an acid and astringent taste and a marked acid reaction. Its salts, having two atoms of its hydrogen displaced by bases, are neutral to test-paper. Except those of the alkali metals, the dimetallic and trimetallic meconates are mostly insoluble in water. Meco- nates are nearly all insoluble in alcohol. They are but slightly or not at all decomposed by acetic acid. Solutions of meconates are precipitated by hydrochloric acid, as explained above. Solution of meconic acid is colored red by solution of ferric chloride. One ten-thousandth of a grain of the acid in one grain of water with a drop of the reagent acquires a distinct purplish-red color ("WORMLEY). The color is not readily dis- charged by addition of dilute hydrochloric acid (distinction from Acetic acid), or by solution of mercuric chloride (distinction from sulphocyanic acid). Solution of acetate of lead precipi- tates meconic acid or meconates as the yellowish- white meconate of lead, Pb 3 (C 7 HO 7 ) 2 , insoluble in water or acetic acld.^ Excess of baryta water precipitates a yellow trimetallic meconate. So- lution of nitrate of silver in excess precipitates free meconic acid on boiling, and precipitates meconates directly, as yellow trimetallic meconate ; if free meconic acid is in excess, the preci- pitate is first the white dimetallic meconate : both meconates being soluble in ammonia and insoluble in acetic acid. Solution of chloride of calcium precipitates from solutions of meconic acid, and even from neutral meconates, chiefly the white mono- metallic meconate, CaH 4 (C 7 HO 7 ) 2 .2H 3 O, sparingly soluble in cold water. In the presence of free ammonia, the less soluble, yellow, dimetallic salt, CaHC 7 HO 7 .H 2 O, is formed. Both pre- cipitates are soluble in about 20 parts of water acidulated with hydrochloric acid. The separation of meconic acid from opium is effected with least loss by precipitating the infusion with acetate of lead (leav- ing the alkaloids as acetates with some excess of lead in the fil- trate). The precipitate is decomposed, in water, with hydro MIDRIA TIC ALKALOIDS. 339 sulphuric acid gas, and the filtrate therefrom is concentrated (and acidulated with hydiochloric acid) to crystallize themeconic acid. The crystals are purified by dissolving in hot water and crystal- lizing in the cold after acidulation with hydrochloric acid. The calcium rneconate, precipitated in concentrated solution by Gregory's process for preparation of morphia, as by the Br, Pharmacopoeial preparation of morphias hydrochloras, is washed with cold water and pressed. One part of the precipitate is dib solved by digestion in 20 parts of nearly boiling water with 3 parts of commercial hydrochloric acid, and' set aside to crystallize the acid meconate of calcium. The crystals are purified from color and freed from calcium by repeated solution in the same sol vent, used just below 100 C., and each time in a slightly diminished quantity. The acid may be further decolorized by neutralizing with potassic carbonate, dissolving in the least sufficient quantity of hot water, draining the magma of salt when cold, dissolving again in hot water, and adding hydrochloric acid to crystallize. MIDRIATIC ALKALOIDS OF THE SoLANACE*;. 1 The Natural Tropeines. The researches of LADENBURG and others (1879-1 88i) place the group of alkaloids distinguished by active dilatation of the pupil of the eye as isomers of ^the common for- mula, C 17 H 23 NOo. These isomers are compounds of isorneric tropines, C 8 li lti NO, each in combination with the same tropic acid, C 9 H 10 Oo (KRAUT, 1863). The union of the basal tropines with tropic acid is shown as follows : C 8 H 1B NO -|- C 9 H io 3 = C 17 H 23 N0 3 + H 2 O. This synthesis of atropine is realized experimentally. The al- kaloids themselves are termed tropeines. The separation of tropic acid is of the nature of a saponification, being most easily effected by alkalies, as given in full under Atropine, d. The natural midriatic alkaloids having the common formula have been named, partly from their sources in plants, as Atropine, Daturine, Hyoscyamine, Hyoscine, Duboisine, and, probably as C 17 H 23 NO 4 , Belladonnine. Ladenburg reduces these alkaloids in identity to the three isomers, Atropine, Hyoscyamine, and Ilyoscine (beside belladonnine). The alkaloids of the midriatic group, like the Aconite group of alkaloids and like Cocaine, are to be treated with a clear understanding of the fundamental fact cornmcn to these groups, that they are saponifiable bodies a fact that sheds most welcome light upon the long standing difficulty of preserving these alkaloids intact during operations for their authorities classify the midriatic plants under the order of Scroph ulariaeea3. 340 MIDRIATIC ALKALOIDS. separation. * Tlie tropeine alkaloids, including artificial tropines like Homatr opine, have strongly marked chemical characteris- tics including a quite direct relationship to benzene, as shown by " the odor test " ; and an exceptionally strong alkalinity, as shown by phenol-phthalein. The sources of the three natural tropeine isomers already known are as follows : Medicinal drug, and commer- cial alkaloid. Plant. True alkaloids (LADENBUBG). Belladonna, root, leaf. " Atropine." " Heavy atropine " of Merck. " Heavy daturine." Atropa Bella- donna. Atropine (larger part). Hyoscyamine. In root, belladonnine. Total, 0.3 to 0.5 per cent. Root | more than leaves. Hyoscyanms, leaf, seed. " Hyoscyamine." "Light atropine" of Merck. Hyoscyamus niger. H. albus. Hyoscyamine. Hyoscine. Total, 0.1 to 0.5$ Strammonium, leaf, seed. " Daturine." , "Light daturine." Datura Stram- monium. Hyoscyamine. Atropine (a little). Total, 0.2 to 0.3$ Duboisia. " Duboisine." Duboisia mio- poroides. Hyoscyamine. 1 KRAUT, 1863-65: Ann. Chem. Phar., 128, 280; 133, 87; Watts' s Diet., 5, 895. LADENBURG, in part with MEYER, SMITH, and others, 1879 to 1884. A summary to 1883 in Liebig's Annalen, 217, 74; Jour. Chem. Soc., 1883, Abs., 670; Proe. Am. Pharm., 32, 316; Am. Jour. Phar., 55, 463. Jour. Chem. Soc., 1884, Abs., 761. Tropine, the common base of the Atropine group of alkaloids, is a deriva- tive of pyridine. Pyridine, C 5 H 6 N, is the primary member of the pyridine se- ries, C n H 2n - sN, and* the type of the quinoline series, CnH 2n _ , ,N. Both series have great interest in the chemistry of natural alkaloids, many of which are found to be clearly placed in the aromatic group. There is a great deal of evidence now making it probable that alkaloids generally are hydrogenized derivatives of pyridine. See under Cinchona Alkaloids, Constitution of, and Quinoline. (LADENBURG, 1884, 1885; A. W. HOFMANN. 1884, 1885; HANTZSCH, 1884; KONIGS, 1884. Review in Am. Chem. Jour., 1882-85: 4, 64. 157; 5, 60, 72; 7, 200. Ladenburg places the rational formula of tropine, as C 6 H 7 (C 2 H4 . OH)N(CH 3 ) = C 8 Hi 5 NO. That is, in a fe^rahydro-pyridine (C 5 H 9 N), ethylene-hydroxyl (C 2 H 4 .OH) and methyl (CH 3 ) take the place of H 2 . And then atropine, or tro- pate of tropine, stands as C.Ht(C*H4.WC 9 H,03)N(CH,) = C 17 H 23 N0 3 . The comparison with Aconite Alkaloids and Cocaine, mentioned in the text, is extended by the fact that they all yield either benzoic acid or a deriva- tive of benzoic acid, by saponification the tropic acid from atropine saponifica- tion easily changing, through atropic acid, to benzoic acid. PITURINE. 341 In belladonna root an alkaloid other than the "atropine" of commerce was found by HUBSCHMANN in 1858, and named Bella- donnine. LADENBURG (1884) finds belladonnine to be, probably, the tropate of oxytropine, C 17 H 2 3NO 4 , an oxy-atropine. POEHL (1876) found the "daturine" of strammonium to be optically levo-rotatory r while the " atropirie " of belladonna was inactive, and several observers have stated that the " daturine " is the stronger of the two in physiological effect. Hager's Commen- tar (1883) asserts that the former reputation of " English atro- pine" for superiority arose from its having been made from strammonium, " daturine " being more powerful than " atropine." It will be observed that Ladenburg places the difference between " atropine '' and " daturine " chiefly in the larger proportion of pure atropine in the former, and of pure hyoscyamine in the lat- ter. Also the medicinal " hyoscyamine," the total alkaloid of the hyoscyamus, is generally reported to have more intense physiological action than the " atropine " taken as total alkaloid of belladonna (DUQUESNEL, 1882). Undoubtedly these differ- ences in effect are covered by greater differences of strength, due to incomplete separation from impurities not alkaloids. The Duboisia Hopwoodii, dried leaves and twigs of which constitute the Australian drug " Piturie," yields an alkaloid quite different from duboisine of D. mioporoides, a liquid, vola- tile alkaloid, not containing oxygen, resembling nicotine, not midriatic, and named Piturine (LIVERSIDGE, 1881). (See under Piturine.) For Atropine, with the reactions, estimation, etc., of the midriatic group of alkaloids, see p. 344 ; for Hyoscine, p. 342 ; Hyoscyamine^ p. 342; Hoinatropine (one of the artificial alka- loids of the midriatic group), p. 343. PITURINE. CglTgN. 1 From the Duboisia Hopwoodii of Australia. The dried leaves and twigs of this plant consti- tute the drug piturie, used by the natives, with effects chiefly the same as those of tobacco. The yield of alkaloid is stated to be one per cent, of the dried drug. Piturine is a liquid, volatile alkaloid, of oily consistence, slightly heavier than water. It has an acrid, burning taste, a tobacco-like odor, and gives the physio- logical effects of tobacco. Its reaction is strongly alkaline, and it neutralizes acids. It is soluble " in all proportions " of water, alcohol, and ether. It gives precipitates with the general reagents LIVERSIDGE, 1881: Phar. Jour. Trans. [3] n, 815; Am. Jour. Phar., 53, 352. The literature of this alkaloid is chiefly dependent upon the report of this author. 342 MIDRIA TIC ALKALOIDS. for alkaloids, and differs from nicotine in its reactions with mer- curic chloride, gold chloride, platinic chloride, and in Palm's test for nicotine with hydrochloric and nitric acids. HYOSCYAMINE. C 17 H 23 NO 3 = 289. An isomer of Atropine (LADENBURG), which it closely resembles (see p. 344). For sources and relations see p. 340. It will be observed that the distinct alkaloid hyoscyamine forms a small part of manufactured medicinal u atropine," a large part of " daturine," an especially large part of " light daturine," the whole alkaloid of " duboisine," and one of the two alkaloids of the " hyoscyamine " of the mar- ket (the mixed alkaloids of Hyoscyamus niger), the other alka- loid of this drug being Hyoscine. The article sold as " crystal- lized hyoscyamine " is stated to consist mainly of true hyoscya- mine, while " amorphous hyoscyamine " consists chiefly of the alkaloid hyoscine. "Hyoscyamine" is presented in the Ph. Fran, as free alkaloid, in crystalline form, with the " observa- tion" that the article of commerce is commonly amorphous. There appears to be some indirect evidence that hyoscyamine (true) is a more potent midriatic than atropine (see Atropine, l>\ but the greater activity of " hyoscyamine," as total alkaloids of Hyoscyamus niger, is mainly due to the hyoscine, which is a more active midriatic than either atropine or hyoscyamine. Hyoscyamine crystallizes in slender, colorless needles, which sometimes radiate in groups. It melts at 108 C. (236.4 F.) Its solubilities are nearly the same as those stated under Atropine, c. It responds to the distinctive tests given under Atropine. d, with the differences there stated for the reactions with gold chloride, platinum chloride, mercuric chloride, and picric acid. In treat- ment for tropine, with alkalies, a tropine isomeric with that of atropine, and a tropic acid identical with that of atropine, are obtained. Hyoscyamine is well separated from atropine, and less easily from hyoscine, by the precipitation with gold chloride (see un- der Hyoscine). The separation from, liyoscyamub leaf and seed, and methods of quantitative estimation, are given under Atro- pine, e and f. HYOSCINE. C 17 H ?3 NO 3 = 289. Tropate of pseudotropine. An isomer of Atropine (LADEHBUKG), one of the two alkaloids obtained from Hyoscyamus niger, leaf and seed, p. 340. It has been stated that the so-called " amorphous hyoscyamine " of the market has consisted mainly of hyoscine. Hyoscine hydro- bromide, and other salts, presented as such, in crystalline form, HOMA TROPINE. 343 are offered for sale for medicinal uses. An ordinary dose by the mouth is ^ grain, a full dose 1 grain ; it is a calmative, with effects distinct from those of atropine ; its midriatic effects are more rapid than those of atropine ; J obtained by a quantity smaller than required of atropine. 3 Hyoscine, uncornbined, is in syrupy or amorphous solid state, colorless, forming, with ordinary acids, solid salts. The hydro- bromide crystallizes in prisms without color ; the hydriodide in pale golden prisms. The hydriodide is levo-rotatory. In solu- bilities hyoscine resembles Atropine. Hyoscine responds to the distinctive tests given under Atropine, d. The hyosoine auro- chloride is less soluble and less lustrous than the hyoscyamine aurochloride ; it crystallizes in yellow prisms and melts at 198 C. (the aurochloride of atropine, lustreless, nearly insoluble precipi- tate, melts at 185 C. ; that of hyoscyamine, lustrous golden, melts at Io9 C. LADENEURG). The platinochloride of hyoscine forms small octahedral crystals, soluble in water and in alcohol (of hyoscyamine, triclinic crystals ; of atropine, monoclinic crys- tals). Iodine solution in potassium iodide gives a dark-colored, oily product. Potassium ferrocyanide gives a white, amor- phous precipitate. The precipitate with Mayer's solution is yellowish ; with mercuric chloride, amorphous, sometimes oily liquid. Treated with barium hydrate for tropine, as given under Atropine, d. the isome^ named pseudotropine is obtained. This crystallizes in rhombohedrons, melts at 106 C., and boils at 241 C. (tropine melts at 62 C.), dissolves readily in water and in chloroform, sparingly in ether. The aurochloride melts at 198 C. The tropic acid of hyoscine is identical with that of hyoscyamine and atropine. LAI/ENEURG separates hyoscine from hyoscyamine by forma- tion of the aurochloride, which is crystallized several times for removal of the more soluble hyoscyamine salt (atropine auro- chloride, if present, being removed at first as a nearly insoluble precipitate). The crystals obtained are decomposed with hydro- gen sulphide, the filtrate made alkaline with potassium carbonate and shaken with chloroform, and the resulting chloroform solu- tion evaporated to give a residue of the hyoscine. HOMATROPINE. C 16 H 21 NO 3 . Phenyl glycollic tropeine. One of a group of artificial alkaloids called tropeines, and produced by 'H. C. WOOD, 1885: Therapeutic Gazette, 9, 1, 594, 760. 2 By one-fifth the quantity (!) EMMEET. See also HIRSCHBEEG, 1881. 344 MIDRIATIC ALKALOIDS. LADENBUKG (1880) by uniting tropine, the common base of natu, ral atropine and hyoscyamine, with various acidulous and other radicals, p. 339. Homatropine is formed by the union of tro- pine, C 8 H 15 KO, with mandalic acid, C 8 H 8 O 3 , a molecule of water being separated. Mandalic acid is formed from amygdalin by digestion with hydrochloric acid, and in other ways, and has the structure C 6 H 5 . CHOH . CO 2 H, phenyl-gly collie acid. When mandalate of tropine is digested with hydrochloric acid, the ele- ments of water are withdrawn and homatropine is produced. C 8 H 15 NO.C 8 H 8 O 3 =:Ci 6 H 21 NO 3 + H 2 O. Since about 1882 homatropine hydrobromide lias been used medicinally. It is an active midriatic ; its effects do not continue as long as those of atropine, and in the same doses it is less poisonous. Homatropine is crystallizable in prisms from a solution in absolute ether ; has a melting point of about 98 C. ; is hygro- scopic and very deliquescent ; and it is ordinarily obtained only in the state of a thick liquid. It dissolves some in water, but is not freely soluble in water. It is freely soluble in ether and in chloroform. The nydrobromide, C 16 H 21 NO 3 .HBr, crystallizes in flat, rhombic prisms forming wart-like aggregations, perma- nent in the air. 1 It is soluble in ten parts of water, the solution not readily suffering change. The hydrochloride is very soluble in water, and is crystallizable. The sulphate crystallizes in silky needles. With solutions of homatropine salts potassium mercuric iodide gives a white, curdy precipitate ; gold chloride, a pre- cipitate, C 16 H 21 NO 3 . HC1 . AuCl 3 , at first of oily consistence, soon crystallizing in prismatic forms ; picric acid, a precipitate soon becoming crystalline. Platinic chloride gives a precipitate only in concentrated solutions, but fine crystals of the double salt are formed (with the hydrochloride). ATROPINE. C 17 H 23 NO 8 = 289. Tropate of tropine, and pro- bably C 5 H 7 (C 2 H 4 O . CO . CHC 6 H 5 .CH 2 ,OH)NCH 3 (LADENBURG). For sources and chemical structure see p. 339. Forms the larger part of pharmacoposial " atropine " ; a smaller portion of the "daturine" or "atropine" obtained from strammonium; and an isomer of the alkaloids hyoscyamine and hyoscine. Atropine and its isomers (hyoscyamine, hyoscine) are identified as midriatics by the organoleptic test (b, d) ; as tropine tropates by Vitali's test, the crystallizable bromine precipitate, the phenol- 'F. B. POWER, 1882: a summary upon homatropine, Am. Jour. Phar., 54, 145. A TROPINE. 345 phthalein reaction, the test with mercuric chloride, and (if in suf- ficient quantity) by the odor tests (d). Atropine and its isomers are distinguished from each other by the precipitation with gold chloride, and by differences in reactions with mercuric chloride, platinum chloride, and picric acid, that with gold chloride being serviceable for separation. Hyoscyamine from atropine by their melting points. Atropine and its isomers are separated from crude drugs, extracts, plasters, animal tissues, the urine, etc., by general and special methods given under e and are estimated ^n quantity, by gravimetric, volumetric, or physiological method, as laid down under f. Tests for impurities, g. a Colorless or white, lustrous acicular crystals, or a crystal- line or nearly amorphous powder. In commerce sometimes yel- lowish. By exposure to air it acquires at length a yellowish or even violet tint. Melts at 114 0. (237.2 F.) (LADENBUKG, 1881) (U. S. Ph.) At 115-115.5 C. (239-240 F.) (E. SCHMIDT, 1880). The medicinal atropine heated alone on a bath of glyce- rine begins to melt at about 104 C. and is entirely melted at 113 C. (SQUIBB, 1885). The artificial alkaloid melts at 113.5 C. (LADENBUKG, 1883). At 123 C. gives a faint mist of micro- scopic sublimate, not crystalline (BLYTH). Vaporizes at about 140 C,, giving white fumes and an oily sublimate. 1 Vaporizes slightly with boiling water, and even with boiling alcohol (DuA- GENDORFF). When dry does not lose weight at 100 C (DUNSTAN and RANSOM, 1886). Upon ignition it is easily dissipated, with- out residue. b. Without odor, it has a disagreeably bitter and acrid taste. The largest medicinal dose is about -fa grain (Ph. Germ.), and it is an active deliriant poison. Solutions for application to the eye should never exceed the strength of one per cent., and for the test of midriasis should be far more dilute than this. The cat is a favorable subject for the test. A solution in 130000 parts of water, applied to one eye of a cat, suffices for dilatation (DRAGENDORFF). With frogs a solution of 1 to 250 obtains dila- tation, commencing in about five minutes (v. GKAEFE). Dr. E. R. SQUIBB (1885 s ) reports the following results upon the human eye, on applying to one eye of each person one drop of a solu- 1 As to the form of the crystals, formed under the microscope, see HELWIG, 1864; A. PERCY SMITH, 1886. *Ephemeris, 2, 855. See also " Blyth on Poisons," 1885, New York, p. MIDRIATIC ALKALOIDS. tion of atropine sulphate diluted nearly in the proportions here stated : Dilution. Individuals under trial. Commencing dilatation. 2280 parts Several. 15 to 18 minutes. 4560 " Several. 30 minutes. 9120 " Two. About 40 minutes. 18240 u Two. 50 and 32 minutes. 45600 ' " Two. 45 minutes each 91200 " Five. In two, no effect \ in three, effect in about 1 hour. The same persons were not used in all experiments, and eighteen persons in all were employed. The sulphate of atro- pine was " about the best obtainable in the market." A sample of crude atropine fresh from an assay of belladonna leaf gave earlier dilatations, and in the last trial, by dilution to 90800 parts, gave dilatation in 45 and 50 minutes. The investigator esti- mated that an effect in 50 minutes was obtained by action of about 0.000000427 gramme of the alkaloid. Atropine is ex- creted by the urine to some extent, being found in that fluid after administration. In frogs the constitutional effects of atropine are peculiar, including first paralysis, and, after a day or later, tetanic spasms. c. Atropine is soluble in 600 parts of water at 15 C. (59 F.), or in 35 parts of boiling water ; soluble in glycerine, and freely soluble in alcohol, chloroform (3 parts), ether (60 parts), amyl alcohol, and benzene (42 parts), scarcely soluble in petroleum benzin or carbon disulphide. Fixed oils dissolve it. Aqueous solutions are not very stable. It has a decided alkaline reaction, exhibited not only with litmus-paper in common with most alka- loids, but with phenol-phthalein, a difference of atropine and its isomers from other alkaloids (FLUCKIGEK, 1886). Its special alkalinity is also shown by its reaction with mercuric chloride (see d). Its salts with the stronger acids are freely soluble in 1 GRAEFE gives 1 to 28000 as the dilution for moderate dilatation commenc ing in | to 1 hour. A TROPINE. 347 water or alcohol ; not soluble in chloroform or ether. At 50 to 60 C. both benzene and amyl alcohol extract a little atropine from acidulous solution (DRAGENDORFF). The sulphate crystal- lizes anhydrous, (C 17 H 23 NO 3 ) 2 H 2 SO 4 . The salicylate is of neu- tral reaction, not crystallizable, deliquescent, not stable in solu- tion. Dr. SQUIBB (1885) advises the preservation of aqueous so- lutions of atropine sulphate by salicylic acid, a cold saturated solution of which is taken for one half of the solvent. d. In evidence of the presence of atropine, the physiological test for the pupil-dilating alkaloids, chiefly atropine and its iso- mers (p. 340), deserves to be named first. Of bodies other than the solariaceous alkaloids it is to be observed that cocaine, digi- talis and its active principles, and conine dilate the pupil of the eye. Aconitine has a variable effect of dilatation. Nicotine is stated to first dilate and then contract the pupil. SELMI (1877- 1879) found certain ptomaines to dilate the pupil. The visual effect of the solanacese seems to have been imperfectly known prior to the last quarter of the eighteenth century. 1 The limits are given under b. Dilatation from v solution not stronger than 1 in 500 parts causes little inconvenience to the human eye. The eye of the cat is preferable. In testing the separated product of an analysis, an aqueous solution is obtained of the free alkaloid or its salt (sulphate), neutral or only very feebly alkaline in reac- tion, and not strongly saline with any metallic salts, and not al- coholic. A drop or two is let fall into one of the eyes, the time noted, and from time to time the width of the one pupil is com- pared with that of the other. VitaWs test is made as follows : The dry residue is treated with a little fuming nitric acid, then dried on the water-bath, and when cold touched with a drop of solution of potassium hydrate in absolute alcohol, when, in evidence of atropine (or one of its isomers), a violet color will appear, slowly changing to a dark red. Strychnine gives a red, brucine a greenish color. The violet color is distinctive for atropine among all important alka- loids, and reaches the limit of 0.000001 gram of the alkaloid (D. YITALI, 1880). ARNOLD (1882) in this test uses, instead of fum- ing nitric acid, first a drop of sulphuric acid rubbed, cold, to moisten the residue, and then a solid particle of sodium nitrite. With atropine the violet does not appear till the alcoholic potash is applied (strychnine, orange-red). Colors appearing before the 1 See an interesting historical paper by ROBERT, 1886: Therapeutic Gazette, 10 445. 348 MIDRIA TIC A LKA L OIDS. alcoholic potash is added (narceine, morphine, narcotine) render the test inapplicable. Phosphomolybdate of sodium gives a yellow precipitate, visible in dilution to 4000 parts (DRAGENDORFF), dissolving in ammonia with a blue color. Iodine in potassium iodide solu- tion, better applied to the hydrochloride solution, gives a precipi- tate of the color of the iodine solution, oily at first and afterward crystalline (LADENBURG), distinct in solution of 8000 parts of water and visible in solution of 50000 parts (JORGENSEN), more complete than precipitation by phosphoniolybdate (DUNSTAN and RANSOM), dissolved by boiling alcohol, from which solvent it crystallizes, blue-green, as pentahydriodide. For separation of the alkaloid from this precipitate see e. Bromine dissolved to saturation in hydrobrornic acid solution gives a yellow precipi- tate, at first amorphous, obtained in a solution of the alkaloid in 10000 parts of water (WORMLEY '), not dissolved by acids or fixed alkalies. The amorphous precipitation is common to most alka- loids, but the precipitate of atropine and its isomers is character- istic in this (Wormley) that it soon becomes crystalline, and under a magnifying power of 75 to 125 diameters presents distinctive forms of lanceolate leaf-like crystals, which gradually group to- gether like the petals of a flower. These crystals may be obtained from spontaneous evaporation of one grain of a solution diluted to 25000 parts. Imperfect crystallization gives only irregular needles and granules. Repeated trials are made by dissolving in a drop of water and crystallizing anew. Phenol-phthalein as an indicator applied to the free alkaloid, as to the chloroformic or ethereal residue, gives the scarlet color in evidence of alkalinity, this reaction being, according to Prof. FLUCKiGER, 2 common to atropine and its isomers and hoinatro- pine, and a distinction from all alkaloids in general use. Mercuric chloride in a 5 per cent, solution in 50 per cent, alcohol, avoiding an excess, gives a red precipitate containing mercuric oxide (GERRARD, 1884 ; SCHWEISSINGER, 1885 ; FLUCKI- GER, 1886). On standing tabular crystals of atropine mercuric chloride are obtained. With hyoscyamine the precipitate appears only after warming (Schweissinger). With this reagent most alkaloids give white precipitates ; morphine and codeine, yellow ones. Of course inorganic bodies of alkaline reaction must be absent, and the alkaloid must be free. Potassium mercuric iodide, or Mayer's solution, gives a whitish, curdy precipitate, 1 " Microchemistry of Poisons," 3d ed., 1885, 641. *1886: Phar. Jour. Trans. [3] 15, 601; Am. Jour. Phar., 58, 129. A TROPINE. 349 Jiardly visible in solution diluted to 4000 parts (DRAGENDORFF). Potassium bismuthic iodide, a precipitate visible in solution diluted to 25000 parts (THRESH, 1880). Gold chloride, a lustre- less precipitate, discernible in solution diluted to 20000 parts, melting at 135 C. (LADENBURG), C 17 H 24 NO 3 . AuCl 4 . The hy- oscvamine precipitate, with gold chloride, has a golden lustre and melts at 159 C. (LADENBURG). Platinum chloride (with hydro- chloric acid) precipitates only very concentrated solutions of atropine ; the crystals of chloroplatinate are monoclinic and melt at 207 C. (hyoscyamine chloroplatinate crystals are tricliriic) (LADENBURG). Picric acid (HAGER) with the "English atro- pine " (p. 341) gave an amorphous turbidity, which, after heating to dissolve it, crystallizes in rectangular plates on cooling. The " German atropine," treated in the'same way, gave the rectangu- lar plates at once. Tannic acid precipitates neutral and con- centrated solutions of atropine. The dilute caustic alkalies, and sodium and potassium normal carbonates, precipitate, from con- centrated solutions of atropine, a part of the alkaloid, soluble in an excess of a caustic alkali. On heating the fixed alkali solu- tions ammonia is finally evolved by decomposition of the atro-_ pine. Ammonium carbonate and fixed alkali bicarbonates give no precipitates. Concentrated sulphuric acid gives no color. Atropine, in common with its isomers, is easily saponified, or resolved by alkalies into its TROPINE and TROPIC ACID. (See un- der Midriatic Alkaloids.) The aqueous solution of alkaloid is digested with barium hydrate at 60 to 80 C. ; then carbon dioxide is passed in and the barium carbonate filtered out ; the filtrate is acidified with hydrochloric acid and shaken with ether, in two portions ; the separated ether is allowed to evaporate spontaneously, when tropic acid is obtained in the residue. The aqueous solution left after removing the ether is now treated with potassium hydrate solution to an alkaline reaction, and the liquid again extracted with ether, which is allowed to separate after shaking, drawn off, and evaporated in a warm place. The residue will contain the tropine. Instead of digestion with barium hydrate, digestion with hydrochloric acid may be employed. Tropic acid melts at 117 C. Heated with dilute solution of permanganate it gives odor of bitter-almond oil, and on further treatment benzoic acid is formed. Tropic acid is easily changed, by loss of H^O, to atropic acid, C 9 H 8 O 2 , isomeric with cinnamic acid. Tropine crystallizes from anhydrous ether in the rhombic system, and melts at 62 C. It is hygroscopic, in ordinary resi- dues assumes an oily consistence, is freely soluble in water, in alcohol, and in ether, has a strong alkaline reaction, gives an odor 350 MLDRIATIC ALKALOIDS. when heated, and forms definite salts with acids. The chloro- platinate crystallizes with orange-red color, dissolves in water, not in alcohol. The odor test, by production of benzoic or salicylic aldehyde, is made, in several ways, by concentrated sulphuric acid alone, or by this followed by dichromate or other oxidizing agent, and is directed as follows in the Ph. Germ. : " 0.001 grarn [at the least] of the atropine sulphate, in a small test-tube, is heated until white vapor appears, then 1.5 grams of sulphuric acid is added, and heated until it commences to brown. Now on adding 2 grams of water an agreeable odor is perceived, and by further addition of a crystal of permanganate of potassium the odor of bitter-almond oil is obtained." This reaction, by whatever reagents, is not a delicate one, and often fails, but it is characteristic in comparison of ordinary alkaloids. e. Separations. Aqueous solutions of atropine, in concen- tration by heat and in standing, aie liable to suffer very slight waste of the alkaloid by its decomposition, but this waste is less for salts with stable acids than it is for the free alkaloid, and in ordinary evaporations is prevented by adding enough dilute sul- phuric acid to neutralize or barely to acidulate the liquid. Acidulous solutions of atropine can be washed by petroleum benzin without loss of the alkaloid, and washed by chloroform or ether with only so much loss as results from the slight misci- bility of the water with these solvents. Chloroform or ether, preferably the former, or, if separations require, benzene or amyl alcohol, by agitation (in repeated portions) with aqueous solu- tions made alkaline, will extract the alkaloid almost without waste The certainty of complete separation is assured by a negative result in testing the aqueous solution with phosphomo- lybdate, or iodine in potassium iodide, or a residue oil evaporat- ing a portion by Vitali's method. Also, it is important to remem- ber that when an acidulous solution is made alkaline a salt is formed, as ammonium sulphate, and such salt will be carried into chloroform or ether or amyl alcohol, and on evaporation of the solvent a crystalline residue of the salt will be obtained. If the salt be ammonium or other alkali sulphate, the atropine is safely separated from the residue by solution in absolute alcohol. Again, water acidulated with sulphuric or hydrochloric acid, agitated with a solution of free atropine in chloroform or other above-named solvent, gradually transfers the alkaloid to the aqueous solution. The remaining chloroforrnic or ethereal liquid is tested, as to the progress of the separation, by subjecting a ATROPlNb. 351 residue from a small portion to Yitali's test. An aqueous solution so obtained may be precipitated by iodine in potassium iodide solution for estimation of the alkaloid, as directed on p. 354. In separating the aqueous layer from an under-layer of chlo- roform, or from an over-layer of ether or benzene, a " separator " made for the purpose is, on some accounts, the most convenient vessel, but the use of a large, strong test-tube, or test-glass on foot, with the wash bottle fittings for siphon-decantation, is very satisfactory. These forms of apparatus are figured and described under Alkaloids, pp. 35, 36. For separations from belladonna root and leaves DUNSTAN and RANSOM (1884, 1885) 1 direct as follows : u Twenty grams of the dry and finely powdered root are exhausted by hot percola- tion with a mixture of equal parts by volume of chloroform and absolute alcohol; and if an extraction apparatus is used about 60 c.c. of the mixture is required. The percolate is agitated with two successive 25 c.c. of distilled water [acidulation having been found unnecessary], which [the watery layers] are separated in the usual way. These are mixed and well agitated with chloro- form to remove the last traces of mechanically adherent coloring matter. The chloroform is separated, the aqueous liquid rendered alkaline with ammonia and agitated with two successive 25 c.c. of chloroform, which are separated, mixed and agitated with a small quantity of water (rendered faintly alkaline with ammonia) to remove adherent aqueous liquid. The chloroform is then evaporated and the residue dried over a water-bath until the weight is constant, which usually occupies a little less than an hour." The alkaloid is obtained in white, silky crystals, for weight, and by trial found pure alkaloid. For the leaves " 20 grams, dried and finely powdered, are well packed in an extrac- tion apparatus, and exhausted with about 100 c.c. of absolute alcohol. The alcoholic liquid is diluted with about an equal volume of water made slightly acid with hydrochloric acid. The chlorophyl, fat, etc., are then removed from the slightly warmed liquid by repeatedly extracting it with chloroform until nothing further is removed by the solvent. The aqueous liquid is made alkaline with ammonia, and the alkaloids extracted by chloro- form, by evaporating which a residue of pure alkaloid is obtained, and dried by heating it at 100 C. until a constant weight is attained." 'From the Root: Phar. Jour. Trans. [3] 14, 623; Am. Jour. Phar., 56, 279. From the Leaves and the Extract: Phar. Jour. Trans. [3] l6 f 237, 238; Am. Jour. Phar., 57,582, 584. Additional report, and discussion by Messrs. GERRARD, REDWOOD, and others, Phar, Jour. Trans., 1886 [3] 16, 777, 786. 352 MIDRIATIC ALKALOIDS. Dr. E. R. SQUIBB a exhausts finely powdered root or leaves of belladonna with alcohol slightly acidulated with sulphuric acid, as follows : 50 grams powdered leaves are moistened with 32 grams alcohol of sp. gr. 0.820 to which about three drops (0.1 gram) of sulphuric acid have been added. Pack in a cylin- drical percolator and exhaust, most readily by a water-pump, with alcohol not acidulated, of which about 300 c.c. will be re- quired. Of the powdered root 50 grams are moistened with 15 to 20 grams of strong alcohol acidulated with three drops of sul- phuric acid, and packed rather lightly unless an efficient pump can be used in the percolation. The percolate is evaporated in a shallow dish over the water-bath until alcohol ceases to be per- ceptible in the vapor; the liquid is diluted while warm with 25 c.c. of water to which one or two drops of sulphuric acid have been added ; the mixture stirred well and transferred to a sepa- rator, rinsing the dish with one or two c.c. of water. Wash the dish with two or three portions of chloroform, about 30 c.c. in all, stirring to take up the residue, transfer the whole to the sepa- rator, acidulate with about three drops more of sulphuric acid, and agitate the whole by active shaking for about five minutes, " not so very vigorous as to emulsify the liquids " to an extent preventing separation afterward. Emulsion occurs in the assay of the leaves, not in that of the root. " If, after standing at rest for an hour, the separation shall not have begun, add three drops more of acid, again agitate for a minute or two, and again set at rest for an hour. K the emulsion still does not begin to sepa- rate, add 10 c.c. more of water and of chloroform, again agitate and set at rest." " The stronger the alcohol used the less of this emulsifying matter is carried into the extract. An alcohol of sp. gr. 0.814 used for exhaustion of the powder never gave any trouble from emulsifying." After the chloroform layer (dark colored) is obtained and drawn off, the aqueous liquid is washed with fresh portions of chloroform, of 10 c.c. each, until the chlo- roformic layer is obtained nearly colorless. The total chloro formic washings are now agitated with 15 c.c. of water acidulated with a drop of sulphuric acid, and after separation by rest the aqueous layer is taken off and added to the main watery solution of alkaloidal sulphate. This is now agitated, in the separator, with 20 c.c. of fresh chloroform and as much as 6 grams of crys- tallized sodium carbonate, the last added in small portions to avoid frothing over,. until a decided alkaline reaction is obtained. After agitation arid rest the chloroformic layer is drawn off into *Ephemeris, 2, 849, Sept., 1885. ATROP1NE. 353 a tared ben>er. A second treatment with 10 c.c. more of chloro- form is made, this chloroform being added to the first, and the tared beaker is set in a warm place for the spontaneous evapora- tion of the chloroform. The beaker is then turned on its side in a warm place, and the residue, sometimes crystalline, sometimes varnish- like, is dried, to weigh as atropine. For Ilyoscyamus leaves either method above given for bella- donna leaves may be employed. ITyoscyamus seeds are first freed from fats by exhausting the powder with petroleum benzin (DRAGENDORFF).' Prollius's fluid, a mixture of 88 parts of ether, i of ammonia-water, and 8 of alcohol, may be used to exhaust the drug in separation of atropine and its isomers. Dragendorff (1876-1877) objects to the use of chloroform upon the acidulous solution of hyoscyamus to remove impurities, on the ground that this solvent takes some alkaloid from acidified liquids. He re- commends, instead of chloroform, benzin, benzene, or amyl alco- hol upon the acid solution, and uses chloroform (or benzene), after making alkaline, to take up the alkaloids. DUNSTAN and RANSOM, in the report quoted on p. 351, give a separation of the alkaloid from alcoholic extract of belladonna leaves as follows : " 1 to 2 grams of the extract are warmed with dilute hydrochloric acid until as much as possible is dissolved. The mixture is filtered, preferably through glass wool or cotton wool, and the residue washed with hot dilute hydrochloric acid until nothing further is dissolved. The acid liquid is then re- peatedly agitated with chloroform, which, when evaporated and dried at 100 C., leaves a residue of pure alkaloid." (Compare SCHWEISSINGER, 1885.) In quantitative separation of the alkaloids from belladonna plasters (the mass of which is usually insoluble in alcohol), a weighed portion of the plaster is macerated in chloroform several hours, with addition of enough ammonia to give an alkaline reac- tion, the plaster-cloth is macerated again, and washed, in chloro- form, then dried and weighed to obtain by difference the weight of plaster-mass taken. The chloroformic liquid and washings, with all the suspended matters liberated from the plasters, are now shaken out (in a separator) with two or three successive por- tions of water acidulated with sulphuric acid, the mixture set at rest, and the aqueous layer drawn off. If the partly emulsified mixture resists separation after standing some hours, it may be warmed to near the boiling point of the chloroform (by immers- 1 Further see DUQUESNEL, 1882: /. Pharm. [5] 5, 131; Jour. Chem. Soc., 1882, Abs. 535. 354 MIDRIA TIC ALKALOIDS. ing the separator in warm water). Also small additions of alco- hol, or of fresh chloroform and of water, may be made by turn- ing rather than by shaking the separator. And a shallow dividing layer of persistent emulsion may be managed by transferring it to a test-tube for treatment with additional chloroform and water, or with a little alcohol. The united portions of watery solution, filtered if not entirely clear, are now made distinctly alkaline by adding ammonia, and shaken out with two or three portions of chloroform, which is drawn off clear. Films of emulsion may be washed with chloroform on a filter previously wet with this sol- vent. The clear chloroformic solution is evaporated in a tared beaker for weight. The residue may be purified by dissolving in absolute alcohol (by which sulphates are left behind with traces of other impurities), filtering and washing with the same solvent, and evaporating the filtrate to dryness. Or the residue from evaporation of the chloroform ^lay be purified (according to DUNSTAN and R.) by dissolving in hydrochloric acidulated water, filtering and washing, precipitating with iodine in potassium iodide, collecting the precipitate, shaking it with sodium thiosul- phate solution to liberate the alkaloid, then shaking out with portions of chloroform as before. In the above process the aci- dulous water solution may be washed with petroleum benzin with advantage, unless the constituents in aqueous solution are such as to form a troublesome emulsion with the benzin. In separation from animal tissues and other matters under an analysis for poisons,* the finely divided material is to be di- gested, with the addition (if need be) of water enough nearly to cover the solids and dilute sulphuric acid to strong acidulation, for an hour or two, at about 70 C. The mixture is now filtered by a filter-pump, or strained, and the residue digested some time with two successive smaller portions of slightly acidulated water, each portion being filtered or strained into the first filtrate, the filters being sparingly washed with slightly acidulated water. To the liquid is now added enough calcined magnesia to neutral- ize the excess of acid, still leaving a distinctly acid reaction, and the whole is concentrated on the water- bath to a thin syrupy con- sistence, stirring to promote evaporation and prevent any drying at the edges. The mixture is now drained into a flask ; the dish is moistened with water and then rinsed with repeated small portions of alcohol into the flask, each alcoholic addition being mixed by gentle agitation of the flask, and alcohol further added 1 Farther see DRAGENDORFF: "Gerichtl. Cheraie." BLYTH: " Poisons," p. 346 (New York edition). WHAKTON and STILLE, 4th ed., 1884 (Amory and Wood), p. 425. WORM LEY: " Poisons," 3d ed., 1885, p. 645. AT RO PINE. 355 to make in all about 3 or 4 volumes to 1 volume of the syrupy liquid. A few drops of dilute sulphuric acid are added, the whole is well shaken and set aside for some hours, then filtered, the filter washed with alcohol, and the filtrate evaporated in a flask to remove all the alcohol. The watery solution, if it be not thin and limpid, is 'diluted with only enough water to ob- tain this consistence. The acidulous liquid is now shaken out with one or more portions of petroleum benzin, or of benzene, or of both, repeating (if the layers separate well) as long as the solvent removes organic matters, and lastly the liquid is well shaken out with chloroform. The benzin and benzene solutions may be examined, if desired, for other substances : the chloroform solution is washed with a little water to which a drop of diluted sulphuric acid is added, and this aqueous solution is added to the liquid under analysis. This liquid is now made alkaline by add- ing ammonia, and shaken out with two or three portions of chloroform. The chloroformic solution is evaporated to dryness in a beaker or assay-flask. The residue is dissolved by warm ab- solute alcohol, in repeated small portions, the alcoholic solution filtered through a small filter into a small foot-glass graduated in c.c., the filter being washed several times with a little of the alcohol. Of the mixed filtrate ^ by volume is evaporated to dryness in a tared beaker, for weight : the ^ is evaporated in one or more small porcelain evaporating dishes, for Vitali's test, taking first two or three drops by a pipette to evaporate in the dish, then repeating the evaporation on the same spot until a concentrated residue is obtained. If this test does not reveal atropine, one edge of the residue in the beaker is carefully sub- jected to the same test, after which the remains of the test are fully wiped out with filter- paper, noting what fraction of the entire residue appears to have been removed. Whatever the result of this test upon either residue, the entire (remaining) residue in the beaker is now dissolved in a small, measured .volume of water. The solution is subjected to the physiological test, which may be made quantitative, and to the bromine test, and other precipitations, working with drop portions on a glass slide, over white or black ground, using a magnifier, comparing crystals under the microscope with those of a known solution of atro- pine. It must be ascertained that none of the solvents gives any residue by evaporation. From the urine atropine may be separated by acidulating with sulphuric acid, washing with chloroform, making alkaline by ammonia, shaking out with chloroform, and proceeding from this point as above directed in treatment of the alkaline aqueous liquid. 356 MIDRIA TIC ALKALOIDS. Atropine lias been recovered from tissues after 2| months' putrefaction (DRAGENDORFF). f. Quantitative. Pure atropine is dried at or below 100 C. and weighed as anhydrous alkaloid. In ordinary methods of sep- aration the isomers hyoscyamine and hyoscine, so far as present in the material, will be included in the estimation, affording a statement of total (midriatic) alkaloid, as atropine. For certainty of the absence of non-alkaloidal matter the separation from the periodide precipitate is a resource (p. 354). The crystalline (or amorphous) residue by evaporation of a chloroformic solution prepared with the precautions already given is clean and suitable for gravimetric estimation. In a physiological valuation, taking Dr. Squibb's results with atropine sulphate (p. 345) as a basis, 1 part of atropine sulphate in about 90000 parts of aqueous solution, or 1 part of free atropine in about 106000 parts of aqueous solution, by application of one drop of such solution to the human eye, should cause a percepti- ble difference of dilatation of the pupil within one hour. For convenience of making the solution, 0.010 gram of the alkaloidal material to be tested may be dissolved to make 10.6 c.c. (for free alkaloid), or 9.0 c.c. (for sulphate), and then 10 c.c. of either of these solutions is to be diluted to 1000 c.c. A single full drop is let fall from a dropping tube directly into the eye while the lids are held open and the head thrown back, this position being maintained without winking for half a minute after the drop is introduced. The trial may be repeated on the same individual and on different individuals. If dilatation be not obtained a stronger solution is to be tried, and trials repeated until the limit of strength for dilatation is obtained (p. 346). Precipitation of atropine by Mayer's solution of potassium mercuric iodide is not very close, but has been used by DKA- GENDORFF 1 both in the gravimetric and volumetric way. In the final volumetric estimation the dilution is to be made such that there shall be 1 part of alkaloid in 350 to 500 parts of solution ; the reagent diluted to one-half Mayer's strength ; and an addition is directed of 0.00005 gram of alkaloid for each c.c. of total liquid. The reagent is to be added very slowly, so that the pre- cipitate shall crystallize and be able to subside. In the volume- tric way the end reaction is found by filtering from time to time a few drops through a very small filter, and adding to this fil- trate a drop of the reagent, then returning this portion to the 1 " Werthbestiinmung," 1874; "Plant Analysis," 1882 (London, 1884). ATROPINE. 357 entire liquid, into which the little filter is rinsed with a few drops of water. Each c.c. of Mayer's (full strength) solution denotes 0.0125 gram of alkaloid as atropine (Dragendorff's ex- periments). 1 In the gravimetric estimation by potassium mercuric iodide, Dragendorff directs precipitation of an acidulated solution of the alkaloid, 1 to 350 or 00, with excess of Mayer's solution. After standing 24 hours the precipitate is collected on a small filter and drained, washed \ntli distilled water, dissolved in alcohol of 90 to 95$, and the alcoholic solution and washings are evaporated in a beaker and the residue dried at 100 C. Of the weight of the residue 44.9$ is atropine. The results are not very close, and will vary with the extent of water-washing of the precipitate. g. Tests of purity. Atropine and its salts, when heated over the flame, should vaporize without residue. It should not have a yellowish tint, and when treated with concentrated sul- phuric acid, or with excess of ammonia, should not be colored. One drop of a solution in 1000 parts of water, on the tongue, should cause a bitter and acrid taste. One drop of a solution in 45000 parts of water, placed in the human eye, should cause di- latation in about 45 (not less than 60) minutes. Free atropine and its salts should correspond to the solubilities stated under c. If 0.001 be dissolved in 1 c.c. of water, and a drop of gold chlo- ride solution be added, a lustreless golden precipitate should be obtained. Further in distinction from hyoscyamine and hyos- cine, note the reactions with gold chloride, and the results by saponification, pp. 343, 349. Prof. E. SCHMIDT (1884) proposes that there should be only two commercial names of the midriatic alkaloids, namely : atro- pine, with a melting point of 115 to 115.5 C. (239-240 F.) (compare p. 345) ; and hyoscyamine, with a melting point of 108.5 C. (227. 5 F.) 1 Only empirical data can be used. And whether this factor of Dragen- dorff be adopted, or one obtained with a solution of atropine, the concentration and other conditions must be held uniform. Before the addition of Mayer's solution ceases to increase the precipitate, a condition of equilibrium is reached, in which an addition of a solution of atropine salt also causes a precipitate. GATHER (Zeitsch. anal. Chem., 8, 477) gave to 1 c.c. of Mayer's solution the value of 0.0193; and Mr. Mayer himself, on obscure theoretical grounds, put the value at 0.0145. The latter figure corresponds to the ratio of 1 atom of mercury to 1 molecule of atropine CnEUsNOsCHl^Hgla. But the gravimetric indications of Dragendorff, given in the next paragraph of the text, correspond to the ratio of 2 atoms of mercury to 1 molecule of atropine (CiTHasNOsHIjaHgla. Further upon this reaction see the author's contribution on "Estimation of Al- kaloids by Potassium Mercuric Iodide," 1880: Am. Chem. Jour., 2,294-304, at pp. 208-300; Chem. News, 45, 114-115; Ber. d. chem. Ges., 14, 1421. 358 OPIUM ALKALOIDS. MINERAL OILS, SEPARATION OF. See FATS AND OILS, p. 274. MORPHINE. See OPIUM ALKALOIDS, p. 362. MYRISTIC ACID. See p. 245. NARCEINE. NARCOTINE. See OPIUM ALKALOIDS. OAK BARK TANNIN. See TANNINS. OLEIC ACID. See p. 246. OLEOMARGARIN. See p. 292. OLIVE OIL. See p. 285. OPIUM ALKALOIDS. Alkaloids in the concrete, milky exudation obtained by incising the unripe capsules of Papaver somniferum. List of Alkaloids, with description of those of minor importance. Chemical Constitution. Grouping by color tests. Classification by deportment with sulphuric acid. MORPHINK : Yield in opium ; analytical outline ; crystallization and heat reac- tions of the alkaloid and its salts (a); physiological effects (5); solubili- ties of the alkaloid and its salts (c); color tests and limits of each, precipi- tations (d)-, separations in general, from tissues in cases of poisoning, with limits of recovery and organs of deposition (e); estimation, gravimetric and volumetric, in opium by the several pharmacopceial and other methods, and in tincture of opium, with authorized standards (/); impurities by the pharmacopoeia! standards in different countries (g). Narcotine : analytical outline, constants, and directions for analysis. Codeine : description and analysis ; tests for purity. Apomorphine : description; tests; impurities. (1) Morphine, C 17 H 19 NO 3 . SERTURNER, 1816. See p. 362. (2) Codeine, C 18 H 21 NO 3 . KOBIQUET, 1832. See p. 388. (3) Thebame, C 19 H 21 NO 3 . THIBOUMERY, 1835. Yield, 0.15- 1.0$. Needles or quadratic plates. Of sharp and styptic taste, and tetanic poisonous effect, fatal in doses smaller than those of morphine. Soluble in 140 parts ether, soluble in benzene, amyl alcohol, and in chloroform, not in petroleum benzin. Of alkaline reaction, and neutralizes acids. Colored by sul- fhuric acid, blood-red turning yellow ; according to HESSE 1872) green to brown-green ; by Froehde's reagent, orange ; by nitric acid, yellow. By boiling dilute acids, converted OPIUM ALKALOIDS. 359 into Thebenine, which is changed by concentrated acid to Thebaioine, both these products being isomers of thebaine, and amorphous. <4) Narcotine, C 22 H 23 NO 7 . DESRONE, 1803. See p. 387. ($) Narceine, C 23 H 29 NO 9 . PELLETIEK, 1832. Yield, 0.02 to 0.1$. Crystallizable, in four-sided prisms or in needles. Of bitter taste, styptic after- taste, and purely hypnotic effect. Sparingly soluble in hot water or cold alcohol, soluble in hot alcohol ; nearly insoluble in ether, or benzene, or petroleum benzin, or (WORMLEY) in chloroform ; moderately soluble in amyl alcohol, not from acid solutions. It is of neutral re- action, but unites with acids, forming crystallizable salts. Colored by sulphuric acid, brown or black to yellow or red ; by nitric acid, yellow ; by Froehde's reagent, yellow-brown to blue. (6) Pseudomorphine, C 17 H 19 NO 4 . PELLETIEK and THIBOUMERY, 1835. Yield, not over 0.02 per cent. Lustrous crystals. Tasteless, and of neutral reaction. Insoluble in water, alco- hol, ether, chloroform, dilute acids, or alkali carbonates; soluble in caustic alkalies or lime solution. Colored by sul- phuric acid an olive green ; by nitric acid, orange-red. Forms acidulous crystallizable salts, difficultly soluble in water or alcohol. HESSE infers that this is the same as Oxy- morphine, formed by action of nitrites on morphine, as stated under Morphine, d. (7) Papaverine, C 21 H 21 NO 4 . MERCK, 1848. Yield, sometimes as high as 1 per cent. Crystalline, in white needles or prisms. In action resembles morphine, but is much feebler. It is slightly soluble in cold alcohol and in ether, freely in hot alcohol, moderately soluble in amyl alcohol, in benzene, and in petroleum benzin, soluble in chloroform both from alkaline and from acid solutions. Does not neutralize even acetic acid. Its salts are very difficultly soluble in water ; its sulphate dissolves in sulphuric acid, but is precipitated on adding water. Colored by concentrated sulphuric acid deep blue-violet, soon fading ; Froehde's reagent, violet to blue ; sulphuric acid, with permanganate added, green turning to gray. Dilute solution not precipitated by phos- phomolybdate. (8) Codamine, C 20 H 25 NO 4 . HESSE, 1872. Forms hexagonal prisms, ban be sublimed ; melts at 121 C. Soluble in tot water, in alcohol, ether, chloroform, benzene. Sulphuric acid with a little ferric salt colors it green-blue, at 100 C. dark violet. 360 OPIUM ALKALOIDS. (9) Laudanine, C 20 H 25 NO 4 . HESSE, 18TO. Crystallizes in fine prisms, sparingly soluble in alcohol or ether, soluble in chloroform or benzene. Sulphuric acid containing ferric salt colors it rose-red, at 150 C. violet-red. In physiologi- cal effect, like thebaine, a tetanic poison, more active than morphine, less active than thebaine. (10) Laudanosine, C 21 H 27 NO 4 . HESSE, 1871. Crystallizable. A tetanic poison to animals. Soluble in ether, chloroform, benzene, and in alcohol, insoluble in water. (11) Meconidine, C 21 H 23 ]$rO 4 - HESSE, 1870. Amorphous. Fu- sible at 58 C., instable. Soluble in alcohol, ether, chloro- form, benzene. Colored red and decomposed by mineral acids. Alkaline in reaction. (12) Lanthopine, C 23 H 25 ]S r O 4 . HESSE, 1870. Minutely crys- talline. Slightly soluble in alcohol, ether, or benzene, freely soluble in chloroform. Does not neutralize acids. Sul- phuric acid with ferric salt does not color it. (13) Protopine, C 20 H 19 NO 5 . HESSE, 1871. Crystallizable. Insoluble in water, freely soluble in ether, sparingly solu- ble in alcohol, chloroform, or benzene. Colored dark violet by sulphuric acid with ferric salt. (14) Cryptopine, C 21 H 23 KO 5 . T. and H. SMITH, 1864. Yield very minute. Hexagonal crystals. A bitter taste and pun- gent, cooling after-taste. Hypnotic and midriatic. Insolu- ble in water, ether, or benzene ; soluble in hot alcohol or chloroform. A strong base. Colored by sulphuric acid, blue ; on adding potassium nitrate, green. (15) Gnoscopine, C^HgeNgOj^. T. and H. SMITH, 1878. In long needles. Slightly soluble in cold alcohol, freely solu- ble in benzene, chloroform, or carbon disulphide. Combines with acids. Colored yellow by sulphuric acid, turning red on addition of nitric acid. (16) Rhoeadine, C 21 H 21 NOfl. HESSE, 1865. Also in Papaver Rhoeas. Crystallizable. Not poisonous. Scarcely soluble in water, alcohol, ether, chloroform, or benzene. Feebly alkaline. Does not form definite salts. Mineral acids dis- solve it with a red color, and with the production of jRhoeagenin, isomeric with rhoeadine, a strong base, neu- tralizing acids. Rhoeadiue is colored olive-green by sul- phuric acid, yellow by nitric acid. (17) Hydrocotarnine, C 12 H 15 NO ? . HESSE, 1871. A constituent of opium ; also produced from narcotine. Cotarnine (C 12 H 13 NO 3 ) is first formed by oxidation of narcotine, then reduced to hydrocotarnine (0. R. A. WRIGHT, 1877 and OPIUM ALKALOIDS. 361 earlier). In monoclinic prisms. Poisonous. Soluble in alcohol, ether, chloroform, and in benzene. Slightly va- porizable. Colors yellow in cold sulphuric acid, red on heating. ^Numerous derivatives of the natural alkaloids of opium are known. A description of one of these, Apomorphine, C 17 H 17 NO 2 , is given at the close of this article. Constitution of Opium alkaloids. 1 Morphine and codeine are closely related to each other and to artificial derivatives of each. Narcotine and narceine also are chemically allied, with cotarnine and hydrocotarnine as derivatives. The relation of both these groups to pyridine, the type of alkaloids, has been shown by v. Gerichten. The immediate structure of morphine and codeine is as follows : Morphine, C 17 H 17 NQ j ^ Codeine, C 17 H 17 NO j The structure of narceine and narcotine : (C0 2 H Narceine, (C 13 H 20 NO 4 ) - CO - C 6 H 2 1 OCH 3 ( OCH 3 (COH Narcotine, (Gj^^CR^O^ - CO - C 6 H 2 1 OCH 3 ( OCH 3 Codeine, therefore, is morphine mono-methyl ether. The corresponding morphine mono-ethyl ether is readily formed. And morphine is artificially converted into codeine (GRi- MAUX, 1881) by treatment, successively, with methyl iodide and fixed alkali. Hydrocotarnine is not only found with narcotine^ in opium, but is producible from narcotine by chemical treat- ment. Among the products of decomposition of narcotine are me- conin, opianic acid, and hemepicacid non-nitrogenous bodies, of which the first-named is found ready formed in opium. These 'C. R. A. WRIGHT and co-workers, 1872-1877: Proc. Roy. Soc.. 20; Jour. Ghem. Soc., 25 to 32; on narcotine and narceine, 29, 461; 28, 573; 32, 525; on morphine and codeine, 25, 150, 504. 0. HESSE, 1872: Ann. Ghem. Phar., Suppl. Bd., 8, 261; Jour. Chem. Soc., 25, 721; 1884: Liebig's Annalen, 222, 203; Jour. Chem. Soc., 46. 613. E. v. GERICHTEN, 1881-83: Ber. d. chem. Oes., 13, 1635; 14, 310; 15, 2179; Jour. Chem. Sc., 40, 110, 445; 44, 221. GRI- MAUX, 1881-83: Ann. Chim. Phys. [5] 27, 273; Jour. Chem,. Soc., 44, 358. CHASTAING, 1881: Compt. rend., 94, 44; Jour. Chem. Sc., 42, 413. A list of opium alkaloids, with a few derivatives, arranged in order of the number of car- bon atoms, is given in the Pharmacographia of F. & H., 2d ed., p. 59. 362 OPIUM ALKALOIDS. three bodies are related in structure, as benzene derivatives, as follows : ( (OCH 3 ) 2 ( (OCH 3 ) 2 C 6 H 2 \ COoH C 6 H 2 \ CO-H | COH ( C0 2 H Opianic acid. Hemepinic acid. Hesse (1872) presented a practical division of the opium al- kaloids into groups, according to their deportment with pure sulphuric acid, as follows : 1. a. Dirty dark-green. Morphine, pseudomorphine, codeine. J Dirty red- violet. Laudanine, codamine, laudanosine. 2. Dirty green to brown-green. Thebaine [?], cryptopine, pro- topine. 3. a. Dark-violet. Papaverine. b. Black-brown to dark-brown. Narceine, lanthopine. 4. Dirty red- violet (of shade different from that of 1 5). cotine, hydrocotarnine. MORPHINE. C^H^TTOo 1 = 285. Crystallized, C 17 H 19 NO 3 . H 2 O = 303. (For structure see p. 361.) In opium, as sulphate and meconate. In all parts of the Papaver somniferum, more particularly in the leaves, stems, and seeds just before maturity. In other species of Papaver. Of dried opium crystallized mor- phine forms from 3 to 20 per cent. : by the U. S. Ph. (1880) 12 to 16, average 14, per cent ; by the Br. Ph. (1885) 10 per cent. (or 9.5 to 10.5 per cent.); by the Ph. Germ. (1882) at least 10 per cent. ; by the Ph. Fran. (1884) 10 to 12 per cent. ; these pharmacopceial standards being further defined according to methods of assay given in each. The U. S. Ph. of 1870 specified for dried opium at least 10 per cent, morphine ; the U. S. Ph. of 1860, for opium not dried, at least 7 per cent, morphine. Since 1848 the U. S. customs' service has required of opium imported at least 9 per cent, of morphine on the moist basis equal to 11J-12 per cent, on a dry basis. And so well has the standard of importation been maintained that, for years before the pharmacopoeia of 1880 came into effect, very little opium in reputable hands in this country contained less than about 12 per cent, of morphine on a dry basis. In 1882, before the present pharmacopoeia was issued, Dr. Squibb reported from 230 cases of opium an average of 12.45 per cent, in the dry state, and from 1 LAURENT, 1847: Ann. Ohim. Phys. [3] 19, 361. WRIGHT, 1877, MORPHINE. 363 191 cases an average equal to 12.35 per cent, in the dry state ; and in the same year Mr. C. "W. Parsons reported the assay of 21 Turk- ish opiums with an average of 15.2 per cent, of morphine in the dried opium. In fact, although opium was much stronger than the national pharmacopoeia required it to be, the pharmacopoeia gave no authority for diluting it. Opium, not powdered, could be diluted only by the grossest sophistication. Additions to powdered opium, if made by reputable persons, were made in accordance with an assay and then declared in the brand of the article as containing a specified percentage of morphine. If made by persons without repute or scruple, the limit of the phar- macopoeia would be little regarded. 1 Morphine is recognized under the microscope by its form as free alkaloid crystallizing from solution (a) ; is identified by chemical tests with ferric chloride, sulphuric and nitric acids, Froehde's reagent, phosphomolybdate solution, etc. (d. p. 365) ; is separated by action of amyl alcohol on alkali solutions, and by other means (e) ; and from viscera, in cases of suspected poisoning, and with reference to the recovery of Meconic acid, as directed. Morphine is usually estimated gravimetrically as free alkaloid (f) ; sometimes volumetrically by Mayer's solution (p. 43), or colorornetrically by iodic acid. The estimation of morphine in opium is indexed under f. The yield of mor- phine in opiums is stated on p. 362. The tests of purity of the alkaloid and its salts are given under g. a. Morphine crystallizes, with one molecule of water, in short, translucent, hexihedral prisms of the rhombic (trimetric) system, or in white, lustrous needles. If a few drops of a warm saturated aqueous solution, or a dilute alcoholic solution, be al- lowed to evaporate spontaneously on a glass slide, characteristic crystalline forms are obtained, to be recognized in analysis by comparison with forms from a known portion of morphine treated in the same way. Too rapid evaporation gives an amor- phous residue. Morphine is permanent in the air and below 100 C., and becomes anhydrous at 120 C. (248 R), losing 5.94 per cent, of the weight of the crystals. According to TAUSCH (1880) the water is expelled at 100 C., though slowly. At 150 C., in the " subliming cell " (BLYTH, 1878), morphine " clouds the upper disc with nebulae ; the nebulse are resolved by high mag- 1 Further, an article by the author "On the Strength of Opium and its Preparations in this Country, as compared with the standards of the Pharma- copoeias of 1870 and 1880," 1883: Pro* Mif.h. State Pharm., I, 48. 364 OPIUM ALKALOIDS. nifying powers into minute dots; these dots gradually get coarser, and are generally converted into crystals at 188 C. ; the alkaloid browns at or about 200 C." Heated on platinum foil the crystals melt, then char, and slowly burn completely away. The crystals are of sp. gr. 1.317-1.326 (SCHRODER, 1880). Morphine solutions are levorotatory : [a] r = 88.04 (Bou- CHARDAT; HESSE, 1875). Crystallization and heat reactions of Morphine Salts. The sulphate, (C 1 7H 19 NO 3 ) 2 H 2 SO 4 .5H 2 O = 758, crystallizes easily in colorless, feathery needles of silky lustre, permanent in the air, losing its water of crystallization '(11. 87$) at 130 C. (at 100 C., Ph. Germ.) The hydrochloride, C 17 H 19 NO 3 HC1 . 3H 2 O = 375.4, forms colorless, feathery needles of silky lustre, or minute white, cubical crystals, permanent in the air, parting with the water (14.38$) at 100 C. (TAUSCH, 1880, Ph. Germ.), at 130 C. (FLUCKIGER, " Phar. Chem.") Morphine acetate holds 3H 2 O, with the molec. weight 399, in a white or faintly yellowish- white, crystalline or amorphous powder, slowly losing acetous vapor in the air. Morphine hydrobromide, with 2H 2 O, crys- tallizes in needles, becoming anhydrous at 100 C. (SCHMIDT, 1877). Morphine hydriodide, 2H 2 O, forms needles and rosettes, and parts with its water at 100 C. (BAUEK, 1874 ; SCHMIDT, 1877). b. Morphine is without odor, and of a bitter taste, more promptly obtained from its soluble salts. It is narcotic, hyp- notic, causing contraction of the pupils, and on animals often producing convulsions and paralysis. The fatal dose is not at all uniform for different species of animals of the same size. The alkaloid undergoes change, in part, while passing through the animal body. 1 Further respecting its deposition in the vari- ous organs and recovery therefrom by analysis, see statements under e. c. Solubilities of the free alkaloid. Morphine, crystallized, is very slightly soluble in cold water (1 to 5000-10000, CHAS- TAING, 1882); soluble in 500 parts of boiling water ; in about 100 parts of ordinary alcohol at 15 C., or 86 parts of boiling ordinary alcohol, or 13 parts of boiling absolute alcohol. The saturated solution in cold absolute alcohol contains one part in 60, and is not precipitated by adding water (Fliickiger's " Phar. Chem.") In different conditions, different quantities of solvent are required, as follows the " nascent " condition being that of liberation from 1 ELusemann's "Pflanzenstoffe," 1883, p. 706. ELIASSOR, 1884. MORPHINE. 365 salt in aqueous solution: the ether, chloroform, and amyl alcohol, water-washed : * Ether, Chloro- form. Amyl. Ale. Benzene. Crystallized 6148 4379 91 8930 Amorphous powder 2112 1977 Nascent state. 1062 861 91 1997 Morphine is not dissolved by petroleum ether. The solvents above-named, immiscible with water, do not take morphine from acidified solutions. Morphine is dissolved somewhat freely by aqueous h'xed alkalies, and by 117 parts of water of ammonia of sp. gr. 0.97. It is a decided base, and neutralizes strong acids. Solubilities of Morphine Salts. Morphine sulphate is solu- ble in 23 parts of water at 15.6 C. (DoTT, 1882) ; in an average of 21 parts of water at 15 C. (COBLENTZ, 1882); in 24 parts water at 15 C. (U. S. Ph.) ; in 0.75 part boiling water (U. S. Ph.); in 702 parts of alcohol at 15 C., or 144 parts of boiling alcohol (U. S. Ph.) Morphine hydrocMoride is soluble in 24 parts of water at 15.6 C. (DoTT, 1882), or 0.5 part of boiling water; in 63 parts of alcohol at 15 C., or 31 parts of boiling alcohol ; not soluble in ether (U. S. Ph.) Morphine acetate is soluble in 2^ parts of water at 15. 6 C. (DoTT, method of digestion, 1882); when freshly prepared in 12 parts of water at 15 C , or 68 parts of al- cohol at same temperature (U. S. Ph.) ; in 60 parts of chloroform (U. S. Ph.) It is decomposed by boiling alcohol, so that, on adding water, free morphine is precipitated (Fliickiger, u Phar. Chem.") Morphine tartrate (normal, 3H 2 O) is soluble in 9.7 parts of water at 15.6 C. (!)OTT, 1882) ; morphine meconate (5H 2 O), in 33.9 parts water at 15.6 C. (the same). The hydro- bromide is soluble, the hydriodide slightly soluble, in cold water. d. The color tests for morphine, when the alkaloid is per- fectly separated, are not extremely delicate, as compared with tests for other alkaloids, and are more than usually liable to error from admixture of non-alkaloidal matters. The test with nitric and sulphuric acids ranks first as a means of distinction. Sulphuric acid itself (strictly free from 'The author. 1875: Am. Chem, 6, 84; Jour. Chem. Soc., 29, 405. 366 OPIUM ALKALOIDS. nitric acid) does not color dry morphine (free from narcotine, papaverine, p. 359), or causes only the slightest reddish colora- tion, unless heat be applied. On the water-bath some shade of purple to brown occurs, later deepening to a brown. Sulphuric acid and cane sugar color morphine purple-red. Minute traces of nitric acid cause a violet to purple color in the cold or on slight warming, and this application of morphine constitutes a very delicate though not distinctive test for nitric acid as an impurity in sulphuric acid. Also the sulphuric acid alone is a delicate test for certain other opium alkaloids as impurities in morphine. Nitric acid alone colors morphine red to orange or reddish-yellow the coloration not being intense. Concentrated sulphuric acid with a very little nitric acid gives a violet color. ERDMANN (1861) employed sulphuric acid with intermixture of one per cent, of nitric acid, sp. gr. 1.25 adding 8 to 20 drops to 1 or 2 milligrams of alkaloidal solid for a violet color. HUSEMANN 1 treated the solid alkaloidal matter with a little sulphuric acid, heated the solution above 100 C., but not as high as 150 C., and then touched with a drop or two of nitric acid of sp gr. 1.2, for a dark violet color. JBARFOED (1881) dissolves the solid alkaloid in concentrated sulphuric acid, two drops for each milligram, and heats above 100 C. for a second or two, and then adds, to a thin layer on the porcelain surface, a minute fragment of potassium nitrate, a red color giving evidence of morphine, a violet-red obtained with a good quantity, and a rose-red when but a little is present. Instead of nitric acid, other oxidizing agents, potassium chlorate, or chlorine water may be used. Of the forms of the test above mentioned tlfe last given one is preferred. But the test without heat can be recommended as follows ; To a quantity not over a few milligrams of the dry residue to be tested, on a white porcelain surface, add a drop of pure sul- phuric acid, and rub with a narrow glass rod for a few minutes, not spreading the acid more than is unavoidable. The point of the glass rod is now touched to nitric acid of sp. gr. 1.42, and drawn across the dissolved residue. A red color, violet if in- tense, rose red if less distinct, and soon paling, is the evidence of morphine. The limit of quantity revealed with a good color by this form of the test is about 0.0005 gram of morphine. 2 Huse- mann gave, as the extreme limit, with heat, 0.00001 gram. ! 1863-1876: Arch. d. Pfiar., 206, 231: Zeilsch. anal. Chem., 15, 103. 2 " Control Analyses and Limits of Recovery," by the author, 1885: Proc. Am. Assoc. Adyanc. Sci., 34, 111 ; Chem. News, 53, 78, et seq. Respecting the color reactions of nitric acid, see further a note by the author, 1876: Am. Jour. Phar., 48, 62. MORPHINE. 367 Without heat there is less danger of error due to extraneous matters. Froehde's reagent 0.001 gram of molybdic acid or mo- lybdate of soda freshly dissolved by aid of heat in 1 c.c. of con- centrated sulphuric acid (Dragendorff), and the solution cooled gives a bright color reaction for morphine, quite delicate, but not very distinctive. A drop of the reagent is applied to the dry alkaloidal residue, not over a few milligrams, on a white porce- lain surface. Morphine gives a blue color, simply that of a cer- tain lower oxide of molybdenum, obtained by deoxidation violet- blue when pale, and changing, through shades of greenish-blue, finally to dark blue. 1 Kauzmann placed the limit of quantity of morphine responding to this test at 0.00005 gram ; Wormley, for blue color, at 0.00007 gram. Unless other reducing agents can be excluded it is unsafe to depend upon this test alone as evi- dence for morphine. The iodic acid test is another application of the reducing power of morphine, which promptly liberates iodine from iodic acid, and in presence of starch gives the blue color of iodized starch. It is generally applied to the aqueous solution of a salt of morphine, a single drop of which is enough. DUPRE (1863) directs to evaporate the morphine with a drop of starch solution to dryness, and when cold to moisten with a solution of iodic acid. WORMLEY states that 0.00007 gram of the alkaloid will give a blue color. With very dilute solutions in considerable quantities, liberated iodine may be sought for by shaking, in a test-tube, with carbon disulphide or chloroform. The reaction has been used for a volumetric method of estimation. This test, carefully applied, is scarcely exceeded in delicacy by any other, and it furnishes a confirmation to affirmative results by other tests, but the presence of morphine should never be declared upon the evidence of this test alone. The chemist should clearly understand that a multitude of reducing agents, inorganic and organic, will liberate iodine from iodic acid. The two tests last above given depend upon the reducing power of morphine. Compared with other non-volatile natural alkaloids, it is a strong deoxidizing agent. To give these two tests any value, non-alkaloidal reducing agents, such as tissue substances, must be strictly removed. In any doubt as to their removal, a control analysis may be instituted, in which like tis- 'FROEHDE, 1866; ALM&N, 186b; KAUZMANN, 1869; NEUBAUER, 1870; DRA- GKNDORFF, 1872. " Note on Froehde's Reagent as a test for Morphia," the au- thor, 1876: Am. Jour. Phar., 48, 59. WORMLEY directs a 3 per cent, solution of molybdic acid in sulphuric acid; BUCKINGHAM, a 7 per cent, solution. 368 OPIUM ALKALOIDS. sues or other matters are treated with the same solvents in the same conditions, and the product subjected to these final color tests, in comparison with the residues liable to contain mor- phine. The capacity of morphine for combination with oxygen renders the alkaloid somewhat instable, though in other respects it is a quite stable alkaloid. Among the oxidation products known are Oxymorphine, C 17 H 19 NO 4 (perhaps Pseudomorphine, HESSE), and Oxydimorphine, C 34 H 36 N 2 O 6 , formed by action of silver nitrite, potassium permanganate, or ferricyanide ; and a body of the composition C 10 H 9 1N~O 9 (CHASTAING, 1882). By hot sul- phuric acid, at 150-160 C., " Sulphomorphid? C 34 H 36 K 2 O 8 S, is formed, a body probably closely related to Apomorphine sulphate. As obtained it is a white, amorphous mass. Nitric acid converts morphine into a resinous body, which, treated with potash, yields methylamine ( ANDERSON) . Gold and silver are reduced from so- lutions of their salts by morphine. Another application of the reducing power of morphine has been made in a test by a drop of ferric chloride followed by a drop of very dilute solution of potassium ferricyanide, when a blue color results from the formation of ferrous salt. It is said that a solution of morphine salt in 10000 parts of water gives a blue color by this operation. According to WOKMLEY, narcotine and brucine give this reduction. The reaction is also a test for Ptomaines (which see). LONG (1878 ') observed a reaction of mor- phine with ammonio-cupric sulphate, giving a green color, per- haps due to reduction. When morphine is treated with concen- trated sulphuric acid (p. 365), and then with potassium chromate, a green color is obtained, due to the reduction of the chromium. Therefore, in the fading purple test for strychnine, morphine, if present in sufficient quantity, gives a green color (not fading). Ferric chloride, as a normal salt, with no free hydrochloric acid, in solution of ordinary reagent strength, gives a blue color with morphine or its salts. The solid residue, while cold, on a white porcelain surface, is moistened with a drop of the reagent, or by touching with a narrow glass rod wet with the reagent. According to WOEMLEY, a good deep color can be obtained with 0.0007 gram (0.001 grain) of alkaloid in solid residue. A solu- tion of morphine must be as concentrated as 1 : 600 in order to give the color. Less delicate than the tests previously given, this test is quite as characteristic as any other. u It is necessary, 1 Chem. News, 38 ; Am. Jour. Phar. , 50, 490. 2 A comparison of the tests by iodic acid. Froehde's reagent, and ferric chloride, applied to morphine, grape-juice, orange-juice, and saliva, was reported byD. BROWN, 1878: Phar. Jour. Trans. [3] 8, 70; Proc. Am. Pharm. , 27, 485. MORPHINE. 369 however, to exclude various organic acids of aromatic composi- tion, including the tannins, phenols, salicylic acid, etc., as enu- merated under Phenol, d 1 According to SELMI (1876) certain cadaver alkaloids give the blue color to ferric salts, as well as reduce iodic acid. But these cadaveric alkaloids did not give the violet color obtained by morphine on treatment with a solution of lead dioxide in glacial acetic acid, evaporating at a gentle heat. The general qualitative reagents for alkaloids all respond to morphine. Phosphomolybdate gives a very nearly complete precipitate of a yellowish -white color, dissolving in ammonia with a blue color. Potassium mercuric iodide, or Mayer's solution, gives a less perfect precipitate, not appearing at all in solutions of 1 to 4000. The precipitate approximates to the com- position (C ;7 H 19 lSrO 3 HI) 4 (HgI 2 ) 3 . 2 Iodine in iodide of potas- sium solution gives a reddish-brown precipitate, immediately visible in one drop of a solution of the alkaloid in 10000 parts of water (WORMLEY) ; under the microscope in a 1 to 100000 solution (SELMI, 1876). On standing, reddish-brown crystals form. The precipitate dissolves in alcohol, in alkalies, slowly in acetic acid. Its distinction, under the microscope, from other opium alkaloids, is given by SELMI, 1876. Potassium iodide gives a formation of needle-shaped crystals, somewhat character- istic, obtained only in quite concentrated solutions. Tannic acid and picric acid give precipitates in solutions not very dilute. Alkali carbonates and bicarbonates precipitate morphine, not soluble in excess of the precipitant. Alkali hydrates give a crystalline precipitate, dissolved by excess of fixed alkalies, and by lime solution, sparingly soluble in excess of ammonia. Crystals of free morphine, obtained by precipitation with a little excess of ammonia, or by spontaneous evaporation of a dilute alcoholic or warm aqueous solution, examined under the microscope (in comparison with known morphine under like treatment), give valuable confirmatory evidence of the identity of the alkaloid (p. 363). e. Separations. Aqueous solutions of morphine are con- centrated on the water-bath without marked loss, but if the concentration require long time, or if the solution be complex, in a quantitative separation, it is better to evaporate under dimin- ished pressure at temperature not above 60 to 75 C. From 1 CHASTAING (1881) claims, from the chemical proportions in which mor- phine unites with fixed alkalies, and other considerations, that this alkaloid is in fact a phenol. 2 The author, 1880: Am. Chem. Jour., 2, 294. 370 OPIUM ALKALOIDS. substances insoluble in acidified water or alcohol these solvents remove morphine in its salts, and hot alcohol may be used to dis- solve out the free alkaloid. Of solvents not miscible with water, amyl alcohol is the most satisfactory for morphine. The acidified aqueous solution may be purified, or freed from other alkaloids, by shaking out with benzene, or chloroform, or ether, and finally with amyl alcohol itself. Then the liquid is made alkaline by adding ammonia, and exhausted of morphine by repeated por- tions of amyl alcohol, or by a continuous liquid-extraction appa- ratus supplied with this solvent. It is to be remembered that amyl alcohol carries with it a little of the aqueous solution, so that the amyl alcohol solution requires water-washing, and a little waste occurs. In separation from the tissues and contents of the stomach, or other matters, in analysis for poisons, 1 the solids are finely divided, in a good-sized evaporating-dish, by playing upon the material with a pair of bright, sharp shears. The divided ma- terial may then be treated as directed under Atropine, p. 354, substituting amyl alcohol for chloroform as a solvent of morphine. Tartaric acid may be used for acidulation instead of sulphuric, to favor the rejection of ptomaines (GuAREScm and Mosso, 1883). If it be analysis for opium constituents, it is to be understood that Narcotine is dissolved sparingly by amyl alcohol applied to acidulous solutions, also sparingly dissolved by benzene applied to alkaline solutions, morphine remaining undissolved in both these cases. Unless morphine be found in more than traces, narcotine is not likely to be recovered with identification. Evidence of opium, in distinction from morphine alone, is more confidently sought through tests for Meconic acid. This acid may be separated from the aqueous liquid, in the course for morphine, if acetic acid be added instead of tartaric acid, for acidulation. The filtered aqueous liquid is treated with lead acetate solution, just sufficient to complete a precipitate formed, and filtered. The filtrate is treated with enough hydrogen sulphide gas to throw down all the lead, then filtered, and the filtrate treated in the course of analysis for the morphine The precipitate first formed on adding the lead acetate is washed on the filter with a little water, carried through the filter-point with a thin jet of water, the lead meconate decomposed by hydrogen sulphide gas, the mixture filtered, the filtrate evaporated, the 1 Toxicology: Taylor on Poisons; Blyth's Poisons; Wharton and Stille. vol. 2, 1884; Dragendorff's " Ermittelung von Giften" and "Organischer Gifte"; Wormley's " Microchemistry of Poisons," 3d edition, 1885. STEUVE, 1873: Zeitsch. anal. Chem., 12, 168. MORPHINE. 371 residue taken up with strong alcohol, this solution filtered and evaporated, the residue taken up with warm water, and tested, with ferric chloride and other reagents, for Meconic acid (which see). The residue from the careful final evaporation of the am vl alcohol solution of morphine which may be divided in several dishes for the tests and for weight as directed in analysis for atropine is examined for its deportment in tests by (1) sulphuric and nitric acids, (2) sulphuric and molybdic acids, (3) ferric chloride, (4) iodic acid, and (5) with phosphomolybdate, as direct- ed for each under d. Also (6) a drop of the warm aqueous, or dilute alcoholic, solution is allowed to evaporate very slowly, under the microscope, for crystals of free morphine, to be recog- nized as stated under a. Other tests may be added. The amyl alcohol used should be examined by evaporating a quantity as large as that used in the analysis, and if any fixed residue be obtained, or if a solution of a supposed residue in acidulated water give reactions with general reagents for alka- loids, then the portion of this solvent to be used must be redis- tilled, after adding a little tartaric acid. To decide any question as to results, a control analysis should be carried in a parallel operation upon tissue material as nearly as possible the same as that under examination for poisons. If the tissue material taken be very troublesome, or if the operator prefer, the first solution from the tissues may be an alcoholic acidulous solution, and the residue from the evaporation of this solution may be taken up by water (and a very little acid). If acetic acid be used, care must be taken that acid reaction with litmus be main- tained. It is better that the temperature of evaporations be kept below 80 C., and that concentrations be hastened by a re- duced air-pressure. The recovery of morphine from the body in cases of fatal poisoning by it is by no means always possible. There are numerous recorded cases of failure of competent chemists to find this alkaloid. In the living body morphine is constantly undergoing decomposition. In the dead body it may suffer de- composition at a very slow rate, though it has been found after standing fourteen months in putrefactive liquids (TAYLOR). It is highly probable that morphine undergoes waste by decom- position during a prolonged analytical separation from tissues. On the other hand, when an analysis is commenced immediately after the introduction of morphine into tissue material, ic can be recovered with less waste than attends some much more 'stable alkaloids, probably because it interposes a less degree of adhesion 372 OPIUM ALKALOIDS. than they. In experiments instituted by the author l it was found that the loss in the immediate separation of morphine, in its smallest recoverable quantities, from an avoirdupois pound of tissues, was not over one hundred times the quantity needed for recognition by the test of Husemann. In further experiments in the progress of the same investiga- tion, 2 when 0.32 gram of morphine was administered to a cat, an analysis commenced 40 minutes afterward, the alkaloid was re- covered for identification from the stomach, the kidneys, the urine, and from the blood, but not from the liver. In four ex- periments for quantitative recovery, using estimation by Mayer's solution, results were obtained as follows : In each instance 0.32 gram in solution was introduced directly into the stomach by a stomach-tube ; and in each instance the stomach, liver, heart, and kidneys were analyzed together. In No. 4, when the animal was killed 30 minutes after the administration, and the analysis be- gun at once, the volumetric result indicated the recovery of 0.25 gram of alkaloid. When the animal was killed 4 minutes after the introduction into an empty stomach, symptoms having mean- time occurred, and the body then left for two days, the final ti- tration indicated the recovery of 0.208 gram. When the animal was allowed to survive the administration for 14 hours, and the analysis then at once commenced, the four organs gave only 0.05 gram of alkaloid. On the repetition of the last experiment, but with a delay of 2 days between the death and the analysis, 0.0485 gram was recovered. f. Quantitative. Morphine is usually dried on the water- bath for weight as hydrate, C 17 H 19 NO 3 . H 2 O = 303, 5.94$ water. ^'Control Analyses and Limits of Recovery," 1885: Proc. Am. Asso. Adv. Sci , 34, 111 ; Chem. News, 53, 78, et seqJ From series, each of four graded trials, by the method of separation substantially as given in the text, and by the qualitative test with sulphuric and nitric acids, the following limits of recovery were fixed for a good color test: From 64 grams of bread, 1 part morphine in 185185 parts. " 64 " tissues, 1 " " 142857 " " 64 " liver, 1 " " 142857 " These " tissues" were membranous, as the coats of the stomach, and con- taining much less fat than the liver. Trial of the volumetric estimation of the recovered morphine, when larger proportions of the alkaloid were taken, indi- cated a much greater and much less consistent loss, as follows: From 128 grams of tissues, 1 part morphine in 19608 parts. " 128 ' r liver, 1 " " 10870 " The experiments were performed by Mr. S. G. Steiner, at the request of the author. 2 Mr. Steiner, with the author, in 1885, unpublished. MORPHINE. 373 (See a.) But it is recommended to dry at a temperature not above 85 C. for the weight of the hydrate, or at near 120 C. for the weight of the anhydrous alkaloid. The last-named tem- perature is sustained by the anhydrous alkaloid without loss of weight. The washing of finely crystallized morphine with satu- rated morphine solutions has been resorted to by Teschemacher and others, as specified further on. Still well (1886) proposes to estimate the meconate of lime left as an impurity in the crys- tallized morphine of an opium assay by dissolving and washing with hot alcohol, on a balanced filter, weighing the dried resi- due, and deducting this weight. Besides this gravimetric determination of the free alkaloid there is no well-established method of estimating morphine. The method next to be named, however, is the volumetric estima- tion with Mayer's solution (see Alkaloids, Volumetric Estima- tion of). The solution is adjusted, if necessary by a preliminary assay, to be of the strength of 1 part alkaloid to 200 parts of the solution, and well acidified with hydrochloric or sulphuric acid (alcohol and acetic acid being always absent in this estimation). Undoubtedly the composition of the precipitate is varied some- what by conditions of concentration and preponderance of mass, as occurs with other alkaloids, but when holding the concentra- tion uniform by a preliminary assay (or more than one) the main conditions are fixed. Degrees of acidulation have little effect (DRAGENDOKFF). The end-reaction is found by the completion of the precipitate. A filtered drop is tested on glass slide over black paper, with a drop of the reagent ; and several of these test-portions rinsed from time to time, with a drop or two of water, into the solution under titration. According to Mayer (1862), and Dragendorff and Kubly (1874), 1 c.c. of Mayer's so- lution indicates 0.020 gram morphine hydrate or 0.019 gram anhydrous morphine. The author has obtained results usually a little too low by use of this factor, and recommends standardiz- ing the Mayer's solution with a solution of pure morphine in acidulated water, in conditions of concentration and temperature fixed for the estimation. 1 The composition of the precipitate is given under d, p. 369. An estimation of morphine, in the volumetric and coloro- metric way, by iodic acid, was given by STEIN (1871), by MIL- 1 DRAGENDORFF, 1874: " Werthbestimmung," p. 87. A. B. PRESCOTT, 1878: Pro. Am. Pharm., 26, 812; Jour. Ghem. tioc., 38, 192. And 1880: Am. Chem. Jour., 2, 301; Jour. Ghem. Soc., 42, 664. The aqueous extract of opium, deprived of morphine, yields to amyl alcohol bodies giving a conside- rable precipitate with Mayer's solution. 374 OPIUM ALKALOIDS. LEE (1872), 1 and by SCHNEIDER (1881), and applied to the assay of opium. Aqueous iodic acid is added to a known weight of (opium) solution, and after the lapse of a few minutes the libe- rated iodine is washed out by shaking with carbon disulphide. The sample color thus produced is then compared with a stan- dard color obtained in the same manner from a solution of mor- phine of known strength, and their intensity equalized by add- ing carbon disulphide to the deeper. This method may prove useful in certain exigencies, as where estimations are habitual and there is nothing present besides morphine to reduce iodic acid. The details of the method as improved by Schneider are given where cited. KIEFFER'S a volumetric estimation of morphine consists in a measure of its reduction of potassium ferricyanide. YENTURINI (1886 3 ) finds this to be the most exact of the volumetric methods. Estimation of Morphine in Opium. Processes of Morphio- metric Assay. 4 The following is the process of the U. 8. Ph., adopted in the Revision qflS&Q: u Opium, in any condition to be valued, seven grams (7) ; lime, freshly slaked, three grams (3) ; chloride of ammonium, three grams (3); alcohol, [sp. gr. 0.820J, stronger ether [sp. 1 Archiv d. Phar. [2] 148, 150; Phar. Jour. Trans. [3] 2, 465; Jour. Chem. Soc., 25, 180. SCHNEIDER, 1881: Archiv d. Phar. [3] 19, 87; Proc. Am. Pharm., 30, 232. 2 L. KIEFFER, 1857: Ann. Chem. Phar., 103, 271. 3 V. VENTURTNI, Q-azzetta chim. ital., 16, 239; Jour. Chem. Soc., 50, 1086. 4 In the text following are given the processes of the pharmacopoeias of the United States, England, and Germany, with commentary upon their pro- visions, in comparison with each other. Also, the detailed process of Dr. Squibb, the directions of Prof. Fliickiger respecting modifications of the method of the Ph. Germ., and the experimental criticisms of Mr. Conroy, Mr. H. Lloyd, Mr. Stillwell, and of Messrs. Wrampelmeier and Meinert, with cita- tions from Portes and Langlois, Prollius, and the Soc. de Phar. of Paris. Of further literature a few references are here added: PERGER, 1884: Jour, prakt. Chem. [2] 29, 97; Jour. Chem. Soc., 46, 1217; Pro. Am. Pharm., 33, 298. PROCTER, 1871: Am. Jour. Phar., 43, 65. ALESSANDRA, 1882: Phar. Jour. Irans. [3] n, 994; Pro. Am. Pharm., 30, 231. A. B. PHESCOTT, 1878: with STECHER, Pro. Am. Pharm., 26, 807; Jour. Vhem. Soc., 38, 191; in 1880, "Report on Revision U. S. Ph.," p. 102; with Moss, 1875: Am. Jour. Phar., 47, 460; with JOSEPH F. GEISLER, 1880: " Morphiometric Methods," Ntw Re- medies, 9, 356. In 1883 : ' On the Strength of Opium," etc., Pro. Mich. State Phar. Assoc., i, 48 ; The Druggist, 6. 1. Method of TESCHEMACHER, 1877: Chem. News, 35, 47. Report of T. J. WRAMPELMEIER and G. MEINERT, Mich. State Phar. Asso., Oct. 14, 1886; Am. Druggist, New York, 15, 203. Report of CHARLES M. STILLWELL, 1886: Am. Chem. Jour., 8, 295. A "Bibliography of the Opium Assay" is in preparation by Mr. A. Van Zwaluwenberg, Ann Arbor, and its publication is promised at an early date. MORPHINE. 375 gr. 0.725], distilled water, each a sufficient quantity. Triturate together the opium, lime, and 20 c.c. of distilled water, in a mortar, until a uniform mixture results ; then add 50 c.c. of distilled water, and stir occasionally during half an hour. Filter the mixture through a plaited filter, three to three and one-half inches (75 to 90 millimeters) in diameter, into a 'wide-mouthed bottle or stoppered flask (having the capacity of about 120 c.c. and marked at exactly 50 c.c.), until the filtrate reaches this mark. To the filtered liquid (representing 5 grams of opium) add 5 c.c. of alcohol and 25 c.c. of stronger ether, and shake the mixture ; then add the chloride of ammonium, shake well and frequently during half an hour, and set it aside for twelve hours. Counterbalance two small filters, place one within the other in a small funnel, and decant the ethereal layer as com- pletely as practicable upon the filter. Add 10 c.c. of stronger ether to the contents of the bottle and rotate it ; again decant the ethereal layer upon the filter, and afterward wash the latter with 5 c.c. of stronger ether, added slowly and in portions. Now let the filter dry in the air, and pour upon it the liquid in the bottle, in portions, in such a way as to transfer the greater portion of the crystals to the filter. Wash the bottle, and trans- fer the remaining crystals to the filter, with several small por- tions of distilled water, using not much more than 10 c.c. in all, and distributing the portions evenly upon the filter. Al- low the filter to drain, and dry it, first by pressing it between sheets of bibulous paper, and afterward at a temperature be- tween 55 and 60 C. (131 to 140 F.) Weigh the crystals in the inner filter, counterbalancing by the outer filter. The weight of the crystals in grams, multiplied by twenty (20), equals the percentage of morphine in the opium taken." The Br. Ph., in the Revision of 1885, adopted the following process of opium assay: " Take of powdered opium, dried at 100 O., 140 grains [9.072 grams]; lime, freshly slaked, 60 grains [or 3.9 grams] ; chloride of ammonium, 40 grains [2.592 grams] ; rectified spirit (sp. gr. 0.838), ether (sp. gr. 0.735), distilled water, of each a sufficiency. Triturate together the opium, lime, and 400 grain-measures [25.9 c.c.] of distilled water in a mortar until a uniform mixture results ; then add 1000 grain- measures [64.8 c.c.] of distilled water, and stir occasionally dur- ing half an hour. Filter the mixture through a plaited filter, about three inches in diameter, into a wide-mouthed bottle or stoppered flask ^having the capacity of about six fluid-ounces [Imp. meas., or 170 c.c.] and marked at exactly 1040 grain- measures [or 67.4 c.c.]), until the filtrate reaches this mark. To 3/6 OPIUM ALKALOIDS. the filtered liquid (representing 100 grains [6.48 grams] of opium) add 110 grain-measures [7.1 c.c.] of rectified spirit and 500 grain-measures [32.4 c.c.] of ether, and shake the mixture ; then add the chloride of ammonium, shake well and frequently during half an hour, arid set it aside for twelve hours. Counter- balance two small filters ; place one within the other in a small funnel, and decant the ethereal layer as completely as practicable upon the inner filter. Add 200 grain-measures [or 13 c.c.] of ether to the contents of the bottle and rotate it ; again decant the ethereal layer upon the filter, and afterwards wash the latter with 100 grain-measures of ether added slowly and in portions. 'Now let the filter dry in the air, and pour upon it the liquid in the bottle in portions, in such a way as to transfer the greater portion of the crystals to the filter. When the fluid has passed through the filter, wash the bottle and transfer the remaining crystals to the filter, with several small portions of distilled water, using not much more than 200 grain-measures [or 13 c.c.} in all, and distributing the portions evenly upon the filter. Al- low the filter to drain, and dry it, first by pressing between sheets of bibulous paper, and afterward at a temperature between 55 a and 60 C. (131-140 F.), and finally at 96 to 100 C. (194 to 212 F.) Weigh the crystals in the inner filter, counterbalanc- ing by the outer filter." The weight represents the quantity of morphine in 100 grains [6.480 grams] of the opium. The process of the PJi. Germ., adopted in the revision of 1882, is as follows : Opium is to be dried at a temperature not above 60 C. Of opium powder 8 grams are to be agitated with 80 trams of water, and after half a day the mixture filtered. Of the Itrate 42.5 grams are treated with 12 grams of alcohol (sp. gr. 0.834-0.830), 10 grains ether (sp. gr. 0.728-0.724), and 1 gram of ammonia water (sp. gr. 0.960), and the mixture set aside, in a stoppered flask, with frequent shaking, for 24 hours, at a tem- perature of 10-15 C. The contents of the flask are then brought upon a small filter, of 80 millimeters (3-J- inches), previously dried and weighed. The crystals recovered from the filtered liquid are washed on the filter with a mixture of 2 grams diluted alcohol (59.8$ to 61.5$ by weight) with 2 grains of water and 2 grams of ether, applying this mixture in two portions. The filter and contents are dried at 100 C. Deducting the weight of the filter, the weight of the alkaloid gives the quantity of morphine in 4 grams of opium. Tlie three pharmacopwial processes of opium assay agree in taking a stated quantity of the filtrate to represent a stated frac- tion of the opium taken, thereby avoiding the washing of the MORPHINE. 377 undissolved residue of opium, and without concentration obtain- ing a solution of a strength desired for crystallization. The quantity of filtrate used, in proportion to the total quantity of liquid taken with dried opium for the mixture filtered, is pro- vided in each of these respective processes, and by the authors of similar processes, as follows : U. 8. Ph. . . . For f of the opium, a vol. of filtrate = f vol. of liquids taken. 'Br. Ph ..... " | " " " 6 Soc. Phar. Paris 1 .... " fg " " =! " " " FORTES and LANGLOIS. 2 CONROY 3 .. . WRAMPEL- MEIER and MEINERT, 1886. 4 ____ " f " " " = so Ph. Germ... " " weight " = i PROLLIUS, " 1877 6 ... 4 <4 _ 48 . 5 <4 , PLUCKIGER, f "~ so^o 1885 6 ... j The U. S. and Br. pharmacopoeias, and, earlier, the Pharma- ceutical Society of Paris, take out of the filtrate an aliquot por- tion of the total volume of liquid introduced into the solution subjected to filtration, making no allowance for the volume of solvents being increased by taking solids into solution. Still earlier Portes and Langlois seem to have made such an allow- ance by the increase of 50 c.c. to 53 c.c. Mr. Conroy (1884) assumes that the " 50 c.c. contain the extractive of 5 grams of opium, equal to about 3 grams in- the moist state [italics added] in which it exists in opium. This, from experiments that I have tried, increases the bulk to 52 c.c." Recently Messrs. Wrampelmeier and Meinert have given (Loo. cit.) report of direct experimentation on the question " whether the total liquid that is, the 70 c.c. of water plus the 1 Societe de Pharmacie, Paris adoption of a modification of the process of Portes and Langlois, 1882: Phar. Zeitung, No. 6, from Jour, de Pharm. d' Al- sace-Lorraine ; Am. Jour. Phar., 54, 598. 2 PORTES and LANGLOIS, 1881: Jour, de Pharm. et de Chim., 1881, 399; New Rem., n, 64; Chem. News, 45, 67. 3 CONROY, 1884: Phar. Jour. Trans. [3] 15, 473. * Proceedings Mi f-h. State Phar. Asso., 4, 127; Am. Druggist, New York, 15, 203. 6 PROLLIUS, 1877: Schweiz. Wochenschr. f. Phar., 1877, 381; Proc. Am. Pharm., 26, 276. FLUCKIGER, 1885: Archiv der Phar. [3] 26 ; Am. Druggist, 14, 149. The Ph. Germ, process was contributed by Pliickiger, who presents, later, a slight modification, noticed in the text further on. 378 OPIUM ALKALOIDS. extractive matter dissolved thereby is really more in volume than 70 c.c. or not." These experiments 1 obtained an average total volume of liquid of but TO. 29 c.c., and, so far as they ex- tend, go to support the rate adopted by the U. S. Ph. The method of the Ph. Germ, and of Professor Fliickiger, in which for |- of the opium a weight of filtrate is taken equal to ||f the 'Following is the original report of the experiments of Messrs. Wrampel- meier and Meinert (loc. cit.) : 7 grams of powdered opium were taken, dried at, 100 C., and transferred to a flask. A flask was used instead of a mortar, in order to avoid loss by evaporation. Three grams of freshly slaked lime and 70 c.c. of water were added, the whole thoroughly mixed and allowed to stand for half an hour. The mixture was then placed upon a filter and (instead of 50 c.c.) the liquid was drained off as much as possible by means of an aspirator. The filtrate was weighed, and its specific gravity taken. In order to determine how much liquid there was left in the opium on the filter, the filter was weighed with the funnel, dried at 100 C to constant weight, and again weighed. By multiplying the loss in weight by the specific gravity of the filtrate, the weight of the liquid left in the opium was found. In the same manner the weight of the liquid left in the macerating flask which could not be brought upon the filter was determined. The weight of total liquid was then found by adding to the weight of the filtrate the weight of liquid left in the opium on the filter, and that of the liquid left in the flask, and from this the total volume i.e., the 70 c.c. plus the extractive matter dissolved thereby was calculated by di- viding by the specific gravity. On working two samples of powdered opium in this way, the volume was found to be in the one case 70.83 c.c., and in the other it was 70.85 c.c. ; where- as, according to Conroy, the volume should be 72.8 c.c. Since the U. S. Ph. directs to take opium in any form, it seemed possible that, if lump opium which contains some moisture be used, the volume of liquid might be increased. A sample of lump opium was taken which contained 11 per cent, of moisture. Seven grams were weighed off, cut into small pieces, and transferred to a flask. Then the lime and 70 c.c. of water were added, the whole thoroughly mixed by means of a stirring rod until a uniform mixture was obtained. The mixture was then allowed to stand for half an hour and finally placed upon a filter. The filtrate was weighed and its specific gravity taken, and the weight of the liquids left in the opium on the filter, and that of the liquid left in the flask, were calculated in the above-described manner. Experiments made with two samples gave the following results: /Specific gravity of filtrate, Per cent, of morphine- Total liquid. Experiment I 1.01270 1.01265 1 O"P 8.3 per cent. 9.04 70.39 c.c. 70. 19 c.c. Experiment II Aver 70.29 c.c. This gave an average increase of 0.29 c.c. Then a very moist lump opium containing 20.7 per cent, moisture was used, and the volume of liquid was found to be, in this case, 70.61 c.c. These experiments, therefore, would seem to prove that the volume of filtrate directed to be taken by the Pharmacopoeia (50 c.c.) is nearly correct. MORPHINE. 379 weight of the liquids used with the dry opium, depends upon dry opium containing | of its weight (62. 5$) of soluble mat- ter. The proportion of soluble matter in opium is, at all events, quite variable*. HERBERT LLOYD' found that when morphine itself is sub- jected to the U. S. Ph. process of assay, it suffers a loss equal to from 0.060 to 0.089 gram on the yield of the 50 c.c., greater or less according to the taking of greater or smaller quantities of morphine. Of course it should be understood that an alka- loid cannot be obtained by a single crystallization, as in all established methods of the morphiometric assay of opium, with- out some loss ; nevertheless the result becomes practically a true one when the quantity of the loss is made to equal an average balance of the quantity of impurity remaining in the crystals weighed. It appears from all evidences to be not improbable that, by the U. S. Ph or Br. Ph. process, the loss of weight of real morphine, whatever its sources, exceeds the weight of impurity with the morphine. The quantity of ammonium chloride introduced into the filtrate is to the quantity of dried opium represented in the filtrate, by the directions of the U. S. Ph. as well as by the process of Portes and Langlois, in the proportion of 3 : 5 ; by the Br. Ph. it is 2 : 5. Mr. Conroy (where cited) reports ex- periments showing that excess of ammonium chloride causes proportional diminution of yield. The truth of this conclusion has been confirmed by Messrs. Wrarnpelrneier and Meinert (1886, loo. tit.) From these observations and those of Lloyd (loc. cit.) it appears that morphine and lime exert a mutual sol- vent action on each other, and that other constituents of opium help to dissolve lime. The more lime the more free ammonia. And both free ammonia and remaining ammonium chloride help to dissolve the morphine. 2 It appears, therefore, that the pro- 1 ISS5: Am. Druggist, New York, 14, 221. 5 " In order to find out whether the morphine is held in solution by the excess of ammonia liberated or by the excess of ammonium chloride, the fol- lowing experiments were made. By calculation it was found that, when 0.202 gram of calcium oxide is in solution, 0.399 gram of ammonium chloride is decomposed. Subtracting this from 3 grams, we find that in this case there is an excess of 2.61 grams of ammonium chloride present in the assay liquor. This amount of ammonium chloride was then dissolved in 50 c.c. of pure water and 0.500 gram of morphine added, and the solution allowed to stand for 12 hours, after which time 0.500 gram of morphine had lost 0.135 gram. The amount of ammonia which would be set free in such assay was also calculated, and a solution of 50 c.c. of pure water containing lhat amount of ammonia was found to dissolve, after 12 hours' standing, 110 gram of mor- phine. Thus it was shown that both ammonium chloride and free ammonia in 380 OPIUM ALKALOIDS. portion of ammonium chloride directed by the Br. Ph. is advis- able. The quantity of free ammonia liberated from the ammo- nium chloride in the filtrate is limited by the slight but vary- ing solubility of the lime. The excess of lime in the primary maceration serves to improve the consistence of the mucila- ginous matters of the opium, favoring solution and filtration. This use of lime in excess, which first holds the alkaloid mor- phine in an alkaline solution, and afterward, in the filtrate, be- comes exchanged for free ammonia, (2NH 4 C1 -)- Ca(OH) 2 = 2NH 3 + CaClg + 2II 2 O), is credited to the plan of MOHE. Whether liberated by lime from ammonium chloride, or added in water of ammonia (as by the Ph. Germ.), at all events free ammonia is employed in separating morphine from its com- pounds, to crystallize on standing, in all methods of morphio- metric assay so far well established in use. The crystallization of the alkaloid requires time. In the Hager-Jacobsen processes crystallization was promoted, and the crystals purified, by the addition of email quantities of ether and benzene, not too much to be taken into solution in the crystallizing liquid. The use of an excess of ether, much beyond ether-saturation, so as to cause an ether layer to rise above the crystallizing liquid, along with the frequent shaking up of the ether with the aqueous liquid in the closed fiask during crystallization, marks an important prac- tical advance in opium assay. This use of ether, introduced about 1881, has been adopted in each of the three pharmaco- poeial processes above given, also in the processes on individual authority, as hereafter presented. By this use of immiscible ether in forcible contact by agitation with the aqueous solution, crystallization is greatly quickened, and purer crystals are ob- tained. The effect of stirring was emphasized in 1877 by Tesche- macher, who says : " The rapid and continuous stirring is most important, as the precipitation of the whole morphine in fine powder is thereby effected, instead of the granular or mamrnil- lated condition so frequently met with." This effect on crystal- line precipitates, in numerous analytical operations, is well under- stood at present. The addition of alcohol, in the crystallizing liquid, is well understood to cause whiter and finer crystals to be obtained, but, unless counteracted with ether or by greater solution exert a distinct solvent action upon the alkaloid. It is therefore probable that by using about 1.000 gram of ammonium chloride instead of 3.000 grams, the amount of morphine held in solution will be greatly re- duced." WRAMPELMEIER and MEINEBT, 1886: Am. Druggist, New York, 15, MORPHINE. 381 concentration, alcohol in proportion to its quantity tends to di- minish the yield of crystals. By the processes of the Ph. Germ, and Professor Fliickiger, alcohol, ether, and aqueous liquid hold the proportions by weight 12 : 10 : 43.5 in the crystallizing liquid. By the Br. Ph. process these proportions are by volume nearly as 4 : 2 J : 41 ; and by the U. S. Ph. process, as 5 : 25 : 50. That' is, for 100 parts ~by weight of aqueous solution, the crys- tallizing liquid contains, in parts by weight, nearly as follows : U. S. Ph. Br. Ph. Ph. Germ. Of Alcohol . . 8 (sp ffr 820) 9 (sp. gr. 0.838) 28 (sp. gr. 0.832) -Of Ether ... . 35 (sp. gr. 725) 35 (sp. gr. 0.735) 23 (sp. gr. 0.726) The directions given by Prof. Fluckiger. in 1885, 1 slightly modified from those of the Ph. Germ., are as follows : u Place 8 grams of powdered opium upon a filter of 80 millimeters {3-J- inches) diameter, and wash it gradually with 18 grams (or 25 c.c.) of ether, the funnel being kept well covered ; force out the last drops of filtrate by tapping the funnel, dry the opium on a water-bath, transfer it to a small flask containing 80 frams of water at 25 C., and shake well repeatedly. After 12 ours pour the mixture on the previously used filter, and collect 42.5 grams of the filtrate in a small flask, to which add 12 grams of alcohol [sp. gr. 0.832], 10 grams of ether [sp. gr. 0.726], and 1 gram of ammonia-water [sp. gr. 0.960], stopper well, set aside .at a temperature of 12tol5C., and shake repeatedly. After 24 hours moisten a new tared filter of 80 millimeters [34 inches] diameter with ether, pour upon it the ethereal layer in the flask, add 10 more grams [14 c.c.] of ether to the latter, and shake well. Again pour the ethereal layer upon the filter. When this has passed, pour the whole contents of the flask upon the filter, and wash the crystals of morphine twice with a mixture of 2 grams of diluted alcohol (sp. gr. 0.892), 2 grams of water, and 2 grams of ether. Dry at a gentle heat, finally at 100C., and weigh, adding the morphine which may still adhere to the inside of the flask." Prof. Fliickiger prefers to weigh the morphine in the flask instead of on the filter. The concentration of the (aqueous) solution set for the crys- tallization of the morphine in an opium assay is very nearly 1 to 10, the same in each of the four processes which have been given -those of the Ph. Germ., Br. Ph., U. S. Ph , and Professor 1 See foot-note on p. 377. 382 p OPIUM ALKALOIDS. Flnckiger. In these processes 10 c. c. of water are taken for each 1 gram of opium, with little addition to alter this proportion, which is nearly retained in the crystallizing liquid. 1 In the pro- cess next to be given, that of Dr. Squibb, the plan of using an aliquot part of the digestive solution is rejected. The undis- solved residue of opium is to be exhausted and washed clean, the total filtrates reaching near 20 c.c. for each 1 gram of opium. The entire solution is to be reduced in volume by evaporation, the washings separately, until brought to about 2 c.c., increased, by transfer rinsing and by ammonia-water, to about 3 c.c. of crystallizing liquid for 1 gram of the opium taken. The ether, of course, is not to be counted as solvent, since it serves as an anti-solvent in all the processes. This greater concentration of volume undoubtedly diminishes the loss 3 due to morphine left in the mother-solution, and increases the gain due to impurities held in the morphine crystals. 3 The relation between this loss and this gain, in opium assay, was mentioned on p. 379. The process of Dr. E. R. tiquibb* published in 1882, is as follows: " Take of opium in its commercial condition 5 10 grams 1 Wrampelmeier and Meinert (loc. cit.) object to the U. S. Ph. direction to triturate and digest in an open mortar, and to measure in a wide-mouthed bot- tle or flask, as liable to cause some concentration by evaporating. Such con- centration of volume interferes with the principle of taking an aliquot part, and tends toward too high results. 2 " About 10 per cent, of the morphine in the opium is retained in the mother-liquor after crystallizing the morphine according to the U. S Ph."- "In order to determine how much of the alkaloid is dissolved in the mother- liquor after crystallizing the morphine, a solution was made to correspond as nearly as possible to the assay liquor, and then a certain amount of morphine was used. The amount of lime (CaO) found to be present in the mother-liquor of the lump opium was 0.202 gram. This amount of lime was taken, slaked with a little water, transferred to the flask, and 50 c.c. of distilled water were added. On adding then 0.500 gram of pure morphine it was found that some of the lime was left undissolved. Therefore, in another trial, a little less calcium ox- ide was used, the 50 c.c. of water and 0.500 gram of morphine added. Then, as in the U. S. Ph. process, 5 c.c. of alcohol, and 25 c.c. of ether, and 3 grams of ammonium chloride were added, and the mixture allowed to stand for 12 hours. The amount of morphine obtained was 0.442 gram, showing that of the 0.500 gram taken 0.058 gram was retained in solution in the mother-liquor." WRAMPELMEIER and MEINERT, Am. Druggist, New York, 15, 203. 3 "The precipitate of morphine obtained by Dr. Squibb's process contains insoluble matter, resinous and other organic matters soluble in alcohol, and meconate of lime, the latter constituting about 25 per cent, of the impurities present. The average amount of the impurities present in the crystals obtained by his process is 8 percent, of the weight of the crystals." CHARLES M. STILL- WELL, Am. Chem. Jour., 8, 306. 4 1882: Ephemeris, I, 14; Jour. Chem. Soc., 42, 666. Further, see WAIN- WRIGHT, 1885: Jour. Am. Chem. Soc., 7, 45. 6 " If of lump opium, every tenth lump of a case should be sampled by cut- ting out a cone-shaped piece from the middle of the lump. Then from the side MORPHINE. 383 (154.32 grains). Put the weighed portion in a flask, or common wide-mouthed vial of 120 c.c. (4 f. oz.) capacity, tared and fitted with a good cork. Add 100 c.c. (3.3 f. oz.) of water, and shake well. Allow it to macerate over-night, or for about 12 hours, with occasional shaking, and then shake well and transfer the magma to a filter, of about 10 centimeters (4 inches) diameter, which has been placed in a funnel and well wetted. 1 Filter off the solution into a tared or marked vessel, then percolate the residue on the filter with water dropped on the edges of the fil- ter and on the residue, until the filtrate measures about 120 c.c. (4 f. oz.), and set this strong solution aside. Then return the residue to the bottle by means of a very small spatula, without breaking or disturbing the filter in the funnel, add 30 c.c. (1 f. oz.) water and shake well, and return the magma to the filter. When drained rinse the bottle twice, each time with 10 c.c. (\ f. oz.) water, and pour the rinsings upon the residue. When this has passed through, wash the filter and residue with 20 c.c. (f f. oz.) of water, applied drop by drop around the edges of the filter and upon the contents. When the filter has drained there should be about 70 c.c (2f f. oz.) of the weaker solution. 2 The filter and residue are now to be dried until they cease to lose weight at 100 C. If any residue remains in the bottle, the bot- tle is also to be dried in an inverted position and weighed. [The weights show the quantity of insoluble matter in the opium.] Evaporate the weaker solution in a tared capsule of about 200 c.c. (6f f. oz.) capacity, without a stirrer, on a water-bath until of the cone a small strip is taken from point to base, not exceeding say half a gram from cones which would average 10 to 15 grams. The little strips are then worked into a homogeneous mass by the fingers, and the mass is then wrapped in tin-foil to prevent drying, until it can be weighed off for assay. When opened to be weighed off it is best to weigh off at once three portions of 10 grams each. In one portion the moisture is determined by drying it on a tared capsule until it ceases to lose weight at 100 C. Another portion is used for the immediate assay, and the third is reserved for a check assay if desira- ble." . . . Opium "should not be dried, but should be weighed for the assay in the condition in which it is found in the market, and in which it is to be dis- pensed." 1 "If the shaking be frequent and active," "the time of maceration can easily be shortened even to three hours." The author of the process states that exceptional opiums give a magma which will not filter, and advises to treat such with ether before the assay, washing in a bottle with 30 c.c. ether, shak- ing well, and washing further with 10 c.c. ether and drying on a filter. 2 " This (120 + 70 = ) 190 c.c. (6 f. oz.) of total solution will practically ex- haust almost any sample of opium. But occasionally a particularly rich opium, or one in coarse powder, or an originally moist opium which has by slow drying become hard and flinty, will require further exhaustion. In all such cases, or cases of doubt, the residue should be again removed from the filter and shaken with oO c.c. (1 f. oz.) of water, and returned, and be again washed as before." 384 OPIUM ALKALOIDS. reduced to about 20 grams (309 grains). Now add the 120 c.c. of stronger solution, and evaporate the whole again to about 20 grams (309 grains). "When cool add 5 c.c. (0.17 f. oz.)of alcohol (sp. gr. 0.820), and stir until a uniform solution is obtained and there is no ex- tract adhering undissolved on the capsule. 1 Pour the concen- trated solution from the capsule into a tared flask of about 100 c.c. (3J f . oz.) capacity, and rinse the capsule into the flask with about 5 c.c. of water used in successive portions. Then 2 add 30 c.c. (1 f. oz.) of ether, and shake well. Add now 4 c.c. (0.133 f . oz.) of water of ammonia of ten per cent. (sp. gr. 0.960), and shake the flask vigorously until the crystals begin to separate. Then set the flask aside in a cool place for 12 hours, that the crystalli- zation may be completed. 3 Pour off the ethereal stratum from the flask, as nearly as possible, on to a tared filter of about 10 centimeters (4 inches) diameter, well wetted with ether. Add 20 c.c. (f f. oz.) of ether to the contents of the flask, rinse round without shaking, and again pour off the ethereal stratum as closely as possible on to the filter, keeping the funnel covered. When the ethereal solution is nearly all through, wash down the edges and sides of the filter with 5 c.c. (0. 17 f . oz.) of ether, and allow the filter to drain with the cover off. Then pour on the remain- ing contents of the flask and cover the funnel. When the liquid has nearly all passed through, rinse the flask twice with 5 c.c. (0.17 f. oz.) of water each time, pouring the rinsings with all the crystals that can be loosened on to the filter, and dry the flask in an inverted or horizontal position, and, when thoroughly dry, weigh it. Wash the crystals with 10 c.c. (J f. oz.) of water applied drop by drop to the edges of the filter. When drained, remove the filter and contents from the funnel, close the edges of the filter together, and compress it gently between many folds of bibulous paper. Then dry it at 100 C. and weigh it. Remove the crystals of morphine from the filter, brush it off, and re- weigh it to get the tare to be subtracted. The remainder, added to the weight of the crystals in the flask, will give the total yield of morphine in clean, distinct, small light-brown crystals." 1 " If this solution should contain an appreciable precipitate, as from rare specimens of opium it will, it must be filtered, and the filter be carefully wash- ed through. Then the filtrate must be evaporated to 25 or 30 grams." 2 "If it has been filtered and evaporated, add 10 c.c. (\ f. oz.) of alcohol and shake well." 3 " If the shaking be frequent and vigorous, 2 or 3 hours' time will be suf- ficient to complete the crystallization; or if it be continuous, half an hour will be sufficient, but as a rule it is better to allow the flask to stand over-night." MORPHINE. 335 As to the tests for purity of the recovered morphine, see g, p. 386. To effect a complete washing of the crystallized morphine without loss, TESCHEMACHEK, in 1877, ' resorted to the use of a saturated aqueous solution of morphine and a saturated alcoholic solution of morphine as washing liquids. The " rnorphiated water" was simply a saturated solution, and contained 0.04 per cent, of the alkaloid. The " morphiated spirit " was prepared by mixing 1 part of ammonia-water, sp. gr. 0.880, with 20 parts of (methylated) alcohol, and digesting a large excess of morphine in this mixture for several days. It contained 0.33 per cent, of morphine. STILLWELL, 1886, a adopts this way of washing the crystallized morphine obtained by Squibb's process. He col- lects the crystals on balanced filter-papers of 4 inches diameter. The ethereal stratum of the crystallizing liquid is poured through the filter, washing out several times with 10 c.c. of ether, rinsing the flask around without shaking it, letting settle for a few minutes, and decanting upon the filter. If the aqueous solution pass on to the filter it is of no importance. The washing with ether is followed, first, by a thorough washing with the " morphiated spirit," then by a thorough washing with the " morphiated water," and, after draining, by two more washings of 10 c.c. each of " morphiated spirit." After draining a few minutes, while the funnel is covered with a watch-glass, two additional washings, each with 10 c.c. of ether, are made. " This will re- move any narcotine which may have been left from the evapora- tion of the ethereal solution at the beginning of the operation. The paper and contents are thus left in a condition to be rapidly dried. Let the filter and its contents stand exposed for a few minutes, and then dry at 100 C. and weigh. Twenty minutes' or half an hour's drying is usually sufficient." The purification of the crystals from meconate of lime and any other matters insoluble in hot alcohol, as used by Stillwell, was stated on page 373. The Estimation of Morphine in Tincture of Opium. The following are the directions of Mr. H. B. PARSONS in application of the U. S. Ph. process of morphine estimation to laudanum. 8 Of the laudanum 75 c.c. are evaporated to dryness on the water- 1 E. F. TESCHEMACHER, Chem. News, 35, 47; Jour. Chem. Soc., 32, 231- 232. 2 CHARLES M. STILLWELL, Am. Chem. Jour., 8, 295. 3 " The Composition of the Laudanum generally dispensed in the State of New York," with report of forty-eight samples, 1883: New York State Phar. Asso., New Rem., 12, 194. 386 OPIUM ALKALOIDS. bath. When cool, 75 c.c. of water are added, together with 3 grams of water-slaked lime. Thorough admixture is attained by trituration at intervals during half an hour. The liquid is fil- tered (from calcium meconate and other insoluble matters), and 50 c.c. of the filtrate (representing 50 c.c. of laudanum) is placed in an assay-flask for treatment. Now add alcohol (sp. gr. 0.820), 5 c.c. ; ether (sp. gr. 0.725 or lower), 25 c.c. ; ammonium chlo- ride, 3 grams. Shake the mixture in the corked flask several times during the first half-hour, and occasionally afterward. After 12 or more hours' standing the crystals are gathered on a small balanced filter, slightly washed with cold water, dried at ^60 C. (140 F.), and weighed. The grams of this weight, multiplied by 2 (for 100 c.c. of the laudanum) and by the specific gravity of the laudanum, equal the per cent, of morphine in the sample assayed. Tincture of opium of the U. S. Ph., 1880, is required to be made from one-tenth its weight of dry opium (powdered opium, g. Impurities. "On adding 20 parts of colorless solution of soda or potassa to 1 part of morphine, a clear, colorless solu- tion should result, without residue (absence of other alkaloids)" (U. S. Ph.) "The watery solution of morphine salt is readily made turbid by addition of potassium carbonate. Ammonia gives a precipitate not sensibly soluble in excess of ammonia or in ether, but soluble both in lime solution and in soda solution " (Ph. Germ.) u Take a small portion of the crystals [free morphine from the opium assay], rub them into very fine powder, and weigh off 0.1 gram. Put this in a large test-tube fitted with a good cork, and add 10 c.c. of officinal lime-water. Shake occasionally, when the whole of the powder should dissolve (absence of nar cotine) " (FLUCKIGEK, SQUIBB). " The lime-water test for the narcotine in the results of the assay is quite sufficient, since no- thing, except coloring matter, is so likely or so liable to be present as narcotine. The only difficulty is to know when the lime water has surely dissolved all it will dissolve. This is facilitated by having a very fine powder, and then good judgment is re- quired to know the value or significance of undissolved residues when they are small" (E E. SQUIBB, 1882). " Morphine yields a colorless solution with cold concentrated sulphuric acid, which should not acquire more than a reddish tint by standing some time" (U. S. Ph.) This test, applied with care, gives good comparative indications of the proportions of narcotine. NARCOTINE. 387 If 0.5 gram of morphine sulphate be dissolved in 15 c.c. of water, with the addition of 5 drops of sulphuric acid, and the solution washed with four or five portions each of 25 c.c. of stronger ether, and the united ethereal solutions be evaporated, the narcotine, if any, will be found in the residue. The amount of this residue, and the intensity of its color under action of concentrated sulphuric acid, furnish comparative evidences of the quantity, or comparative quantities, of narcotine as an im- purity in the morphine salt. NAKCOTINE. C 22 H 23 NO 7 = 413. (For structure see p. 361.) Occurs in opium, in very variable proportions, from 1.3$ to 10.9$. Some samples of French opium do not contain any re- coverable by ordinary methods. T. and H. SMITH found an al- kaloid, aconettine, in the roots of Aconitwni Napellus, which they thought was identical with narcotine. Narcotine is characterized by its deportment with pure sul- phuric acid, and with sulphuric and nitric acids (d) ; distin- guished and separated from morphine by its solubility in ether (c), even from feebly acidulous solutions. It is estimated gravi- metrically or by Mayer's solution (f). Separation from opium (0), from morphine, under Morphine (g\ p. 386. a. Crystallizes from alcohol or ether in colorless, transpa- rent, orthorhombic prisms, or in groups of needles, which melt at 170 C., solidifying again at 130, crystalline if cooled slowly, otherwise amorphous. Above 200 C. it splits into meconine and cotarniiie (MATTHIESSEN and WRIGHT). It is heavier than water, odorless, and forms salts of feeble combining force, mostly amorphous, and acid in reaction. The salts are soluble in water, alcohol, and ether, and are to a greater or less extent decomposed either by addition' of much water or (when the acid is a volatile one) by evaporating the solutions. b. JSTarcotine in solution is bitter, when free nearly taste- less. It acts as a narcotic poison, though only in large doses (from 1.5 to 3.0 grams). c. Narcotine is soluble in about 25000 parts cold and 7000 parts boiling water ; when freshly precipitated by ammonia, in about 1500 parts cold and 600 parts boiling water ; in 120 parts cold and 20-24 parts boiling alcohol (96$) ; in 126 parts cold and 48 parts boiling ether (sp. gr. 0.735) ; in 60 parts acetic ether; in 2.7 parts chloroform ; in 300 parts amylic alcohol ; in 22 parts benzene ; slightly soluble in petroleum benzin, which 388 OPIUM ALKALOIDS. takes up only a trace from alkaline solutions (DRAGENDORFF). Chloroform removes it from acid solutions. d. The alkaline hydrates, carbonates, and acid carbo- nates precipitate narcotine (white, crystalline, insoluble in ex- cess of precipitant). Iodine in potassium iodide gives a pre- cipitate (brown), potassio-mercuric iodide (white amorphous), potassium sulphocyanate (amorphous). The other alkaloid reagents also precipitate narcotine, the precipitates not being characteristic. Concentrated sulphuric acid dissolves narcotine, at first colorless, becoming yellow. Upon heating gently the solution becomes orange-red, then violet to dark blue streaks appear, and finally the mixture assumes an intense violet-red color. If the heat be stopped before the violet-red color ap- pears, the solution becomes cherry-red on cooling (HUSEMANN). If a drop of nitric acid be added to a solution of narcotine in sulphuric acid, a red color appears. FROEHDE'S reagent dissolves narcotine with green color, becoming brown, finally reddish. Concentrated nitric acid dissolves it with a yellow color. In the oxidation of narcotine, by nitric acid or acid chromate, or by permanganate, cotarnine and opianic acid are formed as follows : 22 H 23 NO 7 + O = C 12 H 13 NO 3 (cotarnine) +C 10 H 10 O 5 (opianic -acid). 1 e. Narcotine is obtained, in greater part, from the residue after treating opium with water (see Morphine, /), by extract- ing with dilute hydrochloric acid, precipitating with sodium acid carbonate, extracting the precipitate with boiling 80$ alco- hol, and crystallizing out. It is then purified by washing with cold alcohol and recrystallizing from boiling alcohol. Narcotine may also be extracted from opium by means of ether, and crys- tallizes out on concentrating the ether solution. For the sepa- ration of narcotine from morphine see under Morphine, ^, p. 386. f. Narcotine can be estimated gravimetrically. Also by means of Mayer's solution, of which 1 c.c. precipitates 0.0213 gram of narcotine. CODEINE. C 18 H 21 NO 3 = 299. (Structure, p. 361.) In opium, from 0.1 to 1 per cent. 1 WOHLKR, 1844: Ann. Ghent. Phar., 50, 19. MATTHIESSEN and FOSTER, 1860: Ann. Chem. Phar., Supplement B, i, 330 ; Jour. Chem. Soc. [2] i, 342. ANDERSON, Ed. Phil. Trans. , 20 [3] 359 ; Jour. Chem. Soc. , 5, 266. Also see " Watts's Diet./' ii. 89. CODEINE. 389 Codeine is characterized by its solubilities in water, benzene, and ether (1882: Bull. Sue. Chim. [2] 38, 69; Jour. Cham. Soc., 44, Abstracts, 124. Further, CLOETTA and SCHAER, 1881-82: Jour. Cheni. Soc., 42, 106. 4 o6 PHENOL. OH . SO 3 H (C 6 H 6 SO 4 = 174, monobasic). Phenolsulphonates. Sulphophenates. Phenolsulfosauresalz. There are two phenol- sulplionic acids easy of production and liable to occur in sulpho- carbolates of commerce namely : (1) Phenol orthosulphonic acid, having OH : SO 3 H = 1:2, produced by continued con- tact of phenol and concentrated sulphuric acid in equal parts at ordinary temperatures, and the proper constituent of medicinal sulphocarbolates. (2) Phenol para-sulphonic acid, OH : SO 3 H == 1 : 4, produced by heating the ortho acid. Phenol disulphonic acid, CLH 3 . OH . (SO 3 H) 2 , is formed by heating phe- nol with excess of sulphuric acid. 1 Cresolsulphonic acid and xylolsulphonic acid are formed when crude carbolic acid or other mixtures of cresol and xylol are digested with concentrated sulphuric acid. In preparing phenolorthosulphonic acid equal parts of phenol and concentrated sulphuric acid are mixed, after twenty-four hours water is added, and, in some way, the (unavoidable) free sulphuric acid is removed. This may be done by saturating both the phenolsulphuric and sulphuric acids with barium car- bonate and filtering; or by carefully saturating only the sulphuric acid, so that the filtrate shall precipitate neither a barium salt nor a sulphate ; or by saturating both acids with sodium carbon- ate, evaporating, dissolving the phenolsulphate in alcohol, and crystallizing from the filtrate. The higher the temperature of action of the sulphuric acid, and the greater the excess of this acid, the more of phenolparasulphonic acid will result, and its production is not wholly avoided in any case. The potassium phenolorthosulphonate melts at 240 C., and crystallizes in needles with two molecules of water ; the potas- sium phenol parasulphonate melts above 260 C. and crystallizes, anhydrous, in hexagonal plates. 2 Phenolsulphonates are decom- posed with reproduction of sulphuric acid, by boiling with nitric acid or with hydrochloric acid, and slowly by boiling with water. The nitric acid reacts vigorously, as with phenol, forming nitro- phenic acids. Even in water solution at ordinary temperatures free phenolsulphonic acid suffers gradual decomposition. The metallic sulphocarbolates are all measurably soluble in water, the barium and lead salts included (separation from sul- phates). The sodium salt is NaC 6 H 5 SO 4 . 2H 2 O. The alkali 'Sulphuric acid, HO(S0 3 H)' or HO(SO 2 )''OH Phenolsulphonic acid, HO.C 6 H 4 .S0 3 H [C 6 H 6 S0 4 ] Phenoldisulphonic acid, HO.C 6 H 3 .(SO 3 H) 2 Phenylsulphuric acid, C 6 B 5 O.S0 3 e [C 6 H 6 S0 4 ] '"Watts's Dictionary," viii. 1538. PL A NT ANAL YS/S. 407 sulphocarbolates are soluble in much alcohol (another separation from sulphates). Further, the sulphocarbolates are identified by giving the chief reactions of phenol those with nitric acid, bromine, and ferric chloride and by giving the reactions of sulphates only after decomposing with boiling nitric or hydro- chloric acid. PHYSETOLEIC ACID. See FATS AND OILS, pp. 246, 250. PICRACONITINE. See ACONITE ALKALOIDS, pp. 18, 20. PIT URINE. See MIDKIATIC ALKALOIDS, p. 341. PLANT ANALYSIS. The chemical analysis of vegeta- ble tissues. Phytochemical analysis. Systematic methods of chemical analysis of plants have been presented as follows : FREDERICK ROCHLEDER, M.D., professor of organic chemistry in the Uni- versity of Prague, 1858: Wiirzburg, Germany. English translation by Wil- liam Bastick,' London, 1860: Phar. Jour. Trans. [2] i,562. Same translation revised by Professor John M. Maisch, Philadelphia, 1860: Am. Jour. Phar., 33, 81, et seg. ; reprinted in 80 pages, 1862. Dr. G. C. WITTSTEIN, Miinchen, 1868: " Anleitung zur chemischen Analyse von Pflanzentheilen auf ihre organischen Bestandtheilen," 355 pp., Nordlingen. An English translation, "The Organic Constituents of Plants and their Chemi- cal Analyses," by F. von Mueller, Ph.D., F.R.S., Melbourne, 1878, pp. 332. The " plant analysis " is included in Part II., 49 pages. HENRY B. PARSONS, Ph.C., assistant chemist in the Department of Agricul- ture, Washington, 1880: " A Method for the Proximate Chemical Analysis of Plants," American Chemical Journal, I, 377-391; Am. Jour. Phar., 52, 210; Phar. J'>ur. Trans. [3] 10, 793; Jour. Chem. Soc. (abstract), 38, 754; Her. d. chem. Ges., 13, 1370; Chem. Newx (in full), 41, 256, 267: Year-book of Phar., London, 1880, 5u: Jahres. d. Pharm., 1880, 99; Allen's " Commercial Organic Analysis," London, second edition, 1885, i. 356 (tabulated abstract); Lyons's "Pharmaceutical Assaying," Detroit, 1886, p. 37 (tabulated abstract). Given in full in th" following page*. GEORG DRAGENDORFF, professor of pharmacy in the University of Dorpat, Russia, 1882: "Die qualitative und quantitative Analyse von Pflanzen und Pflanzentheilen,' 285 pp., Gottingen. An English translation by Henry G. Greenish, London, 1884: " Plant Analysis: Qualitative and Quantitative," 280 pp. A French translation in Fremy's " Encyclopedie Chimique," Paris, 1885, tome viii. (from the author, without credit to previous publication). An out- line of Dragendorff's scheme is given in the following pages. Good examples of plant analysis, chiefly according to DragendorfFs scheme, have been presented by Helen C. DeS. Abbott, Philadelphia. In 1884, analysis of "Fouquiera splendens," Proc. Am. Assoc. Adv. Sci., 33, 190; Am. Jour. Phar., 57, 81. In 1885, analysis of "Yucca angustifolia," Proc. Am. Assoc. Adv. Sci., 34, 125 (abstract). Other examples are found in reports of results by Parsons's scheme as fol- lows: John Hoehn, Ann A'/bor, 1882, analysis of "cheken leaves," Contribu- tions Chem. Lab. Univ. Mich., I, 39 (abstract). William Heim, Ann Arbor, analysis of ' Piscidia erythrina," ibid., i, 38 (abstract); Ther. Gazette, 1882, p. 4 o8 PLANT ANALYSIS. 254. Henry Palmer, Ann Arbor, analysis of " Viburnum lentago," Proc. Mich. State Phar. Awe., 3 (1885), 158. Results of plant analyses by Mr. Parsons himself are extant as follows: Analysis of "damiana" (Turnera aphrodisiaca), 1880: New Remedies, 9, 261; Phar. Jour. Trans. [3] II, 271. Of "Eupatorium perfoliatum," 1879: Am. Jour. Phar., 51, 342; Archiv der Phar. [3] 15, 557. Of "Berberis aquii'o- [,83; is., 15, 2745. Of "Usti New Remedies, n, 80; Phar. Jour. Trans. [3] 12, 810. Plant analyses in the lium (var. repens)," 1880r New Remedies, n, 83; Phar. Jour. Trans. [3] 13, 46; Ber. d. chem. Oes., 15, 2745. Of "Ustilago maidis (corn smut)," 1880: Reports of the Department of Agriculture at Washington, for 1880, 1881, 1882, as accredited to Mr. Parsons by the chemist, Dr. Collier. See also an excellent article by Mr. Parsons on "Some Constituents of Plants," 1879: New Reme- dies, 8, 168. PAKSONS'S METHOD FOR THE CHEMICAL ANALYSIS OF PLANTS/ Prefatory. It must be premised that no one method is applicable in all cases, and that the operator will so modify and adapt the proposed processes- as to best attain the truths he seeks. If the present scheme shall serve merely as an example, to be improved upon as discoveries multiply, it will at least have served to stimulate to the more thorough study of a much-neglected yet very important branch of analysis. The student, when first entering upon the study of plant analysis, is perplexed and disheartened, owing to the lack of any elementary treatise in which he may find directions for the quantitative estimation of the various plant constituents. The works of Rochleder and Wittstein, while giving most valuable assistance in the investigation of special constituents and their separation from large quantities of the crude herb, still fail to give clear and practicable directions for the quantitative estimation of each constituent. Von Mueller's latest enlarged edition of Wittstcin's " Plant Analysis " gives a scheme, most excellent in many respects, yet cumbered with tiresome methods of extraction and manipulation, which serve to unnecessarily lengthen the time required for making the analyses, without increasing the accuracy of results obtained. Too many American analyses of plants have been summarized thus: ''The plant contains gum, resin, tannin, a volatile oil, and a peculiar bitter principle, to which may be ascribed its medicinal activity." The foreign journals bring occasionally most excellent examples of accurate examinations of vegetable sub- stances; as instances may be cited the examination of ginger, by J. C. THRESH,** and of ergot, 3 aloes, 4 and other articles by Prof. DRAGENDORFF. To these sources the student must look for his best models (p. 407 and above). 1 The publications of this method are cited above, p. 407. Mr. Parsons disclaimed any aim to originality, in the resources used in the scheme (presented at request of the author of this work), but submitted the plan as an outgrowth of his own experience, in a varied practice of chemical analysis of plants and vegetable tissues. 8 Phar. Jour. Trans. [3] 10, 81, Aug., 1879; Am. Jour. Phar., 1879, 51, 519. *Phar. Jour. Trans. [3] 6, 1001, June 17, 1876; Am. Jour. Phar , 1876, p. 413; 1878, p. 335. 4 " Werthbestimmung," 1874, p. 110. PARSONS' S METHOD. 409 In following the plan now presented, the use of the apparatus for con- tinuous percolation is strongly urged for the extractions with benzene, alcohol, and other volatile solvents. A very simple and inexpensive "extraction appa- ratus " has been described by various American and foreign chemists. 1 "In any convenient water-tight vessel is a worm of block-tin pipe, having an internal diameter of 9 mm. and a length of about 2.5 meters. The lower (external) part of this worm is fitted by an ether-soaked velvet cork to a glass percolator having a diameter of 4 cm., a length of 20 cm. to the constriction, and 5 cm. below. Within this percolator is a smaller tube, flanged at the top and bottom, and suspended by fine platinum or copper wires. This tube has a diameter. of 2.5 to 2.8 cm. and a length of 14 cm. ; the bottom is covered by filter-paper and fine washed linen, 2 tied on by linen thread. The weighed sam- ple of the finely powdered herb is placed within this tube for extraction. A light glass flask, weighing about 30 grams, is fitted by an ether-soaked cork to the outer percolator." Having introduced the solvent into this glass flask, the connections are made secure, and heat is applied by a water-bath to the flask. If the liquid is too slowly volatilized the addition of a little common salt to the water in the bath serves to remove the trouble. Next in importance is the use of a good tared filter. The form originally presented by P. A. GOOCH 3 leaves little to be desired. It may be made by per- forating with fine holes the bottom of air ordinary platinum crucible, and fit- ting it accurately to a perforation made in a large rubber cork ; this cork con- nects it with a receiving vessel, which in turn is connected with a Bunsen's pump. Fine asbestos suspended in water is poured into the crucible, the air exhausted from the receiving vessel, and thus a firm, thin layer of asbestos is deposited on the bottom of the crucible. After ignition and weighing, the cru- cible is ready for the reception of any precipitate which it is desired to separate and weigh. The use of these two pieces of apparatus will eliminate two grave sources of error, viz., incomplete extraction of soluble matters, and inaccuracies intro- duced by the use of tared paper filters. The other necessary apparatus is simple, and includes one or more plati- num crucibles and evaporating dishes, accurate burettes and graduated cylin- 'B TOLLENS: Zeitsch. anal. Chem.., 17, 320 (1878): New Remedies, 7, 335, Nov., 1878. W. 0. ATTWATER: Proc. Am. ( hem. Soc., 2, 85 (illustrat- ed). S. W. JOHNSON: Am. Jour. Sci., 13, 196. H. B. PARSONS: New Remedies, 8, 293 (illustrated), Oct., 1879. F. SOXHLET, 1879: Ding, polyt. Jour., 232, 461; Zeitsrh. anal. Chem., ig, 365. MEDICUS, 1880: Zeitsch. anal. Chem., 19, 163; New Remedies, 9, 167 (illustrated). 2 In place of the linen and filter-paper may be substituted fine brass or plati- num wire gauze. Asbestos suspended in water may then be poured in to form a fine felt. The tube can then be dried and weighed, and the amounts ex- tracted may be found by the loss of weight of the tube and substance. A little experimentation will show the operator how to prepare and use the tube. It is but an adaptation of the Gooch's Filter here recommended. s Proc. Am. Acad. Sci., 13,342 (1878); New Remedies, 7, 200 (Oct., 1878); Am. Chem. Jour., i, 317 (illustrated). 410 PLANT ANALYSIS. tiers, a good balance sensitive to at most 0.0005 gram, and the ordinary glass and porcelain-ware found in all laboratories. It is assumed that whoever attempts the analysis of a plant is informed as to the normal constituents to be sought, and that he has had considerable expe- rience in inorganic analysis and in the identification of the principal classes of proximate constituents which he now undertakes to estimate quantitatively. Accordingly, tests for identification will not be here presented ; they should, however, never be omitted. The necessity of recording in detail all physical and chemical peculiarities with every weight that is taken is self-evident. I. Preparation of Sample. The air-dry specimen should be carefully examined, and all extraneous substances removed. The entire sample should then be ground, or beaten in an iron mortar, until it will all pass through a sieve having from 40 to 60 meshes to the linear inch. After thoroughly mixing this sample, take of it about 100 grams, which should be further pulverized until it will all pass through a sieve having from 80 to 100 meshes to the linear inch. From this smaller portion remove all iron, derived from mill or mortar, by use of a magnet. Then place in a clean, dry bottle, which should be labelled and securely corked. This small sample is for the analysis ; the larger portion should be reserved for the sepa- ration of those proximate principles which seem, from the analy- sis, to be worthy of more extended investigation. II. Estimation of Moisture. Dry rapidly, at 100 to 1 20 C., two or more grams of the sample ; the loss of weight equals moisture and occasionally a, little volatile oil. In some cases it is best to dry at a lower tem- perature, and at other times the drying should be conducted in a stream of hydrogen or carbonic anhydride. 1 III. Estimation of Ash. In a weighed crucible gently ignite two or more grams of the sample until nearly or quite free from carbonaceous matter ; the heat should not be permitted to rise above faint redness, or loss of alkaline chlorides may occur. Weigh this residue as crude ash, and in it determine : a. Amount Soluble in Water. This portion may contain chlorides, sulphates, phosphates, and carbonates of potassium and sodium ; also slight amounts of chlorides and sulphates of cal- cium and magnesium. 1 On treatment of fresh plants for drying, and on methods of powdering, see Dragendorff s " Plant Analysis," London edition, p. 6. PARSONS' S METHOD. 411 b. Insoluble in water; Soluble in Dilute Hydrochloric Acid. The residue from a should be treated with a slight ex- cess of hydrochloric acid, and evaporated in a porcelain dish over a water- bath until all free acid has been expelled ; it should then be again moistened with hydrochloric acid, water added, and be filtered from any remaining insoluble substances. This treatment removes carbonates (with decomposition) and phos- phates of calcium and magnesium, sulphate of calcium, and ox- ides of iron and manganese. c, Insoluble in Water j Insoluble in Dilute Hydrochloric Acid j Soluble in concentrated Sodium Hydrate. Boil the resi- due from b with a solution containing about 20 per cent, of sodium hydrate. This treatment removes combined silica of the ash. The residue still insoluble is sand and clay which adhered to the specimen ; this residue should be separated, washed tho- roughly, and weighed. Always determine the amounts removed by the above treat- ment by weighing the dried, undissolved residues. The ash, as thus estimated, usually includes a little unconsumed carbon, to- gether with more or less carbonic anhydride, most or all of which was not originally present in the plant, but was produced by the combustion of the organic matter. For most purposes it is unnecessary to estimate and exclude from the ash this carbonic anhydride ; where great accuracy is desired, a complete quantita- tive analysis should be made, the amount of each base and acid being determined, and in the statement of results only those should be included which existed originally in the plant. For this purpose it is necessary to burn from 20 to 100 grams of the sample ; for further directions consult text books on agricultural and ino-ganic analysis. IV. Estimation of Total Nitrogen. In half a gram or more of the sample determine total nitro- gen by combustion [p. 230 or 229]. If later in the analysis no other nitrogenous substances are discovered, calculate the total amount of nitrogen to albuminoids by multiplying by 6.25 [or 6.33]. When other nitrogenous compounds are present, their content of nitrogen should be determined directly or by diffe- rence ; after proper deductions have been made, the remaining nitrogen should be calculated to albuminoids. V. Estimation of Benzene Extract. In a suitable extraction apparatus completely exhaust 5 grams of the sample with pure coal-tar benzene (sp. gr. 85-88, boil- 412 PLANT ANAL YSIS. ing at 80 to 85 C., leaving no residue when evaporated). The extraction requires from four to six hours' continued action of the solvent. Carefully evaporate this liquid to dryness in a weighed dish, and record its weight as total benzene extract. This extract may contain volatile oils and other aromatic com- pounds, resins, camphors, volatile or non-volatile organic acids, wax, solid fats, fixed oilfi, chlorophyll, other colors, volatile or fixed alkaloids, glucosides, almost no ash. To the weighed extract add water, again evaporate on the water-bath, and complete the drying in an air-bath at 110 C. In absence of other vaporizable substances the loss of weight approximates the amount of volatile oil. If the presence of a volatile alkaloid is suspected (from a characteristic odor or an alkaline reaction), add a drop of hydrochloric acid to prevent its volatilization. Camphors are partially dissipated by this treat- ment ; hence, when they are present, this evaporation should be dispensed with. Treat now the residue with a moderate amount of warm water, allow to stand until cool, then filter through fine paper by aid of a Bunsen's pump. In half of the aqueous filtrate de- termine total organic matter and ash ; test the remaining half for alkaloids, glucosides, and organic acids by salts of lead, sil- ver, barium, and calcium. Care must be taken not to mistake a slight amount of suspended matter, frequently resinous, for other substances actually soluble in water. The still undissolved residue should be again removed from filters and dishes by solution in benzene, the benzene solution being again evaporated to dryness. Treat this residue with warm, very dilute hydrochloric acid, allow to cool, and filter through paper. The filtrate, should be tested for alkaloids and glucosides. The amount extracted by acid, if any, may be deter- mined by weighing the still undissolved residue. Treat this re- sidue with several considerable portions of 80 per cent, alcohol (sp. gr. 0.8483 at 15. 6 C.), allowing at least an hour for each treatment. Filter through paper and determine by evaporation the matter dissolved ; this usually consists of chlorophyll with one or more resins, which may sometimes be separated by use of petroleum benzin, chloroform or similar solvents. Purified animal charcoal removes chlorophyll and some resins from alco- holic solution, while certain other resins are not removed. If camphors were present in the plant, the greater portion will be found in the alcoholic liquid. The substances undissolved by 80 per cent, alcohol may be fixed oil, solid fat, wax, and very rarely a resin; their separa- PARSONS' S METHOD. 413 tion may be attempted by refrigeration and pressure, or by use of ether, chloroform, etc. Recapitulation (portion soluble in benzene or chloroform). \. Loss by evaporation, with precautions : volatile oil. 2. Soluble in water : alkaloids, glucosides, organic acids. ( Insoluble in water : ) A 77 7 . 7 . t1 3 - Soluble in dilute acids : Al ^ l ^ P osslbl 7 4. f Insoluble in water. J Insoluble in acids. 1 Soluble in 80 per cent, alcohol. Removed by animal char- coal : chlorophyll, some resins. j Not removed by animal ( charcoal : some resins. ( Insoluble in water : } 5. < Insoluble in dilute acids : V Wax, fats, fixed oils. ( Insoluble in 80 per cent, alcohol : ) It is frequently advantageous to extract the plant with petro- leum beiizin (sp. gr. 0.66 to 0.70, boiling at about 50 C., wholly volatile) before treatment with benzene ; by reference to the ac- companying table of comparative solubilities (p. 422) it will be seen that this treatment may serve to separate fixed and volatile oils, and some resins ,and colors, from certain solid fats, wax, other resins and colors. Where benzene of sufficient purity cannot be had, pure chlo- roform is the best substitute. The use of ether is objectionable in this place, as its solvent properties are less distinctly marked than are those of benzin, chloroform, and benzene; in other- words, more plant constituents are sparingly soluble in ether than in the above-mentioned solvents. Consequently many sub- stances which should properly be extracted by 80 per cent, alco- hol will be sparingly dissolved if ether were used, while ben- zene, chloroform, and benzin would have no perceptible solvent action upon them ; tannic acids may be cited as instances illus- trating this point. YI. Estimation of 80 per cent. Alcohol Extract. That part of the plant not dissolved by benzene should be dried at 100 C., and then completely exhausted by 80 per cent. alcohol (sp. gr. 0.8483 at 15.6 C.) ' This requires from 12 to 14 hours' continuous treatment with the solvent. Remove, dry, and weigh any crystals or powder that may separate upon concentrat- ing and cooling the alcoholic percolate. Make the clear liquid 414 PLANT ANAL VSIS. to a definite volume (say 200 e.c.), by adding more 80 percent, alcohol. AN ALIQUOT PART (usually 20 c.c.) of this volume of liquid is evaporated to dryness for weight (//i) of organic matter with ash, the residue then ignited for weight (n) of ash, to find by difference (m ri) the amount (o) of organic matter in VI. ANOTHER EQUAL ALIQUOT PART (20 c.c.) is evaporated to remove all alcohol, treated with water, filtered, and the filtrate and wash- ings evaporated to dryness to find the weight (p) of organic matter and ash that are soluble in water, the residue then ignited for weight (q) of ash that is soluble in water. Then p q = r, the amount of organic matter (in VI.) soluble in water. If the plant contain much sugar or much tannin, it will be desirable to proceed now, in separation by water, as directed further on for " the second way" Otherwise proceed in sepa- ration by absolute alcohol, in " the first way" as follows : THE REMAINING ALIQUOT PART (160 c.c.) of the clear alcoholic liquid should be evaporated carefully to dryness, the residue pulverized and treated with several considerable portions of absolute alco- hol (sp. gr. 0.7938 at 15.6 C.) i A. Soluble in Absolute Alcohol (from the portion by 80$ alcohol). a. Soluble in water. a! . Precipitated by subacetate of lead. Tannin and most organic acids ; some extractives ; some inorganic acids of the ash. Weigh in Gooch's filter, ignite cautiously, and again weigh ; loss equals organic matter precipitated. a". Not precipitated by subacetate of lead. Alkaloids, glucosides, some extractives and colors. Determine weight by difference between a and a f . b. Insoluble in water. b r . Soluble in dilute hydrochloric acid. Alkaloids, glucosides (rarely), some extractives. De- termine weight by difference between b and ~b" . b". Insoluble in dilute hydrochloric acid, b'" '. Soluble in dilute ammonium hydrate. Most acid resins, some colors. Determine weight by difference between b" and ~b"" . b"". Insoluble in dilute ammonium hydrate Neutral resins, some colors, albuminoids (in some seeds). Redissolve in alcohol, evaporate, and weigh. PARSONS'S METHOD. 415 B. Insoluble in Absolute Alcohol (from portion by 80$ alcohol). c. Soluble in water. c'. Precipitated l)y subacetate of lead. Some colors, extractives, albuminoids (rarely), organic acids, and inorganic acids of the ash. Weigh in Gooch's iilter, ignite cautiously, and again weigh ; loss equals organic matter precipitated. c". Not precipitated oy subacetate of lead. Alkaloids, glucosides, glucose, sucrose, some extrac- tives. Determine by difference between c and c' '. Eemove Pb by H 2 S, H 2 SO 4 , Na 2 CO 3 , or other means, and titrate for sucrose and glucose. d. Insoluble in water. d'. Soluble in dilute hydrochloric acid. Some alkaloids and glucosides. Determine by diffeiv ence between d and d". d". Insoluble in dilute hydrochloric acid. Few resins, some extractives and color substances. Dissolve in alcohol, evaporate, and weigh in a tared dish. " The second way": primary division of constituents in VI., by solubility in water. In some cases it may be preferable to use the following method for analysis of the 80 per cent, alcohol extract ; it is more desirable when the plant examined contains a considerable amount of sugars, tannic acid, etc. Alcohol Extract, dilute to 200 c.c. with 80 per cent, alcohol. 1. In 20 c c. determine total organic matter and ash. Then, 2, in 20 c.c. determine total organic matter and anh that are soluble in water, and, by difference, total or- ganic matter insoluble in water, as directed in " the first way." 3. Evaporate the remaining 160 c.c. to dryness, treat with water, filter, and make the filtrate measure 160 c.c. Reserve the insoluble matter on the filter for examination (10). 4. In 20 c.c. of the aqueous solution estimate the tannin. 1 5. Precipitate 20 c.c. by normal acetate of lead, and determine, as before described, the amount of organic matter after drying at 100 to 120 C. This precipitate will contain, if the substances are present in the plant, tannic, gal- lic, and most other organic aci'ls, some colors, rarely albuminous substances, some extractives, and most inorganic acids of the ash. Determine, by differ- ence, the amount not precipitated by this treatment. 6. In 20 c.c. determine in like manner the amount precipitated by basic acetate (" subacetate ") of lead. This reagent precipitates a greater number of acids, colors, and extractives than are precipitated by the normal acetate, hence it is frequently possible to estimate such substances by subtracting the amount precipitated by one reagent from the amount precipitated by the other To the filtrate add a slight excess of dilute hydrochloric acid, boil gently for half an hour, and determine in the liquid total glucose by use of Fehling's solution. i For this estimation methods are given in the article on Tannins in this work. Mr. Parsons advised the gravimetric method of CARPENI (1875: Jahr. der Chem., 9 ft 9: Chem. News, 31, 282). Precipitate by ammoniacal acetate of zinc, use a Gooch's filter, wash the precipitate with very weak ammonia, dry at 120 C., weigh, ignite cautiously, again weigh. The loss by ignition equals tannic acid, in absence of certain interfering substances. 4i6 PLANT ANAL YSIS. 7. Precipitate 20 c.c. by subacetate, exactly as in 6, and use the precipitate as a duplicate to check the amount there estimated. To the filtrate add a very slight excess of solution of carbonate of sodium, filter from the carbonate of lead, wash well with water containing a little alcohol, and in the filtrate esti- mate actual glucose. If the glucose thus found is appreciably less than that in i), subtract it from that amount; this glucose maybe due to the presence in the plant of sucrose or some glucoside. If due to sucrose, the amount of the latter may be found by multiplying this residual glucose by 0.95; if to a glucoside, a fit subject for an extended investigation is presented. The properties, formula, and decomposition products of the newly found glucoside should be carefully studied. 8. Precipitate 20 c.c. with subacetate of lead, as in 6 and 7, employing the precipitate as material from which to separate organic acids, after removal of lead by sulphuretted hydrogen. Acidulate the filtrate with sulphuric acid, add an equal volume of alcohol, allow to stand two hours, filter, wash the precipi- tate with 50 per cent, alcohol, and evaporate the filtrate until all alcohol has been dissipated. Test the acid solution for alkaloids, glucosides, sugars, ex- tractives. 9. Reserve the remaining 40 c.c. for duplicating any unsatisfactory deter- minations. 10. The residue mentioned in 3 as insoluble in water may contain resins, albuminoids (especially from seeds), colors, alkaloids, glucosides. Dilute acids remove alkaloids and some glucosides; dilute ammonia water will remove some resins, colors, and glucoxides. Any still insoluble residue probably con- tains albuminous or resinous substances." YII. Estimation of Cold Water Extract. That part of the plant remaining insoluble after treatment with alcohol should be dried at 110 C. and completely extract- ed by cold water. When the plant contains considerable mu- cilaginous matter this is best removed by placing the sub- stance in a flask or graduated cylinder, and then adding a mea- sured volume of cold water. Allow to macerate, with frequent agitation, for from 6 to 12 hours, then filter through line washed linen, and evaporate an aliquot portion of the solution. In this residue determine total organic matter and ash. This residue usually contains little but gum / in analysis of fruits and fleshy roots, pectin bodies, salts of organic acids, rarely a substance re- sembling dextrin, and small amounts of albuminous substances and coloring matter. Usually the separation of these substances is very difficult. The unevaporated liquid should be used for such qualitative reactions as are necessary to show the nature of the substances extracted. The insoluble residue should be well washed with water, transferred to a crucible, and completely dried at 110 C. This residue should then be weighed. VIII. Estimation of Acid Extracts. The dried residue insoluble in cold water should be trans- ferred to a beaker containing 500 c.c. of water and 5 c c. of con- centrated sulphuric acid (sp. gr. 1.84). Boil for 6 hours on a PARSONS' S METHOD. 417 gauze support, adding water to keep the volume of liquid un- changed ; if the substance be very starchy a longer boiling may be necessary. This treatment will convert starch and its amor- phous isomers to dextro-glucose, and will occasionally remove some salt of an organic acid, with usually traces of albuminous and indeterminate substances. The total amount extracted may be found by washing, drying at 110 C., and weighing the yet insoluble residue, and subtract- ing the weight from the one taken after extracting with cold water. The amount of starch and isomers may be found by determining in a given volume of the acid filtrate the amount of glucose, using Fehling's solution ; the glucose thus found multi- plied by 0.9 equals starch and isomers. The total extract minus starch and isomers equals acid extract not starch. ' This includes a small amount of ash, which may be approximately determined by evaporating and igniting a known volume of the solution. Where it is wished to separate the extracted matter from the sulphuric acid, boil the liquid with an excess of powdered barium carbonate until no acid reaction remains. Filter and evaporate to dry ness. The residue consists chiefly of hydrated dextro- glucose (C 6 H 12 O 6 .H 2 O), with some ash. IX. Estimation of Alkali Extract. Wash well and dry at 110 C. the residue from treatment with acid, and record its weight. Boil this residue for two hours with 500 c.c. of a solution containing 20 grams of sodium hydrate to the liter. Filter through fine washed linen, and wash the residue thoroughly with hot water, alcohol, and ether. Transfer it to a weighed crucible, dry at 110 to 120 C., and weigh the residue as crude fibre and ash this weight subtracted from the previous one shows the total alkali extract. This ex- tract is largely albuminous matter and various modifications of pectic acid, Fremy's " cutose" and various coloring, humus, and decomposition compounds, in small amounts. Most of the ex- tracted substances may be precipitated by excess of an acid with or without the presence of alcohol. X. Cellulose. The crude fibre from IX. should be treated with from 50 to 100 c.c. of U. S. Ph. solution of chlorinated soda and allowed to stand twenty-four hours. If not then bleached white, slightly acidulate with hydrochloric acid and set aside for another day. Filter through fine linen or Gooch's filter, wash with hot water. 418 PLANT ANALYSIS. dry at 110 to 120 C., and weigh, ash-free, as cellulose. The loss of weight by this treatment state as lignose and color. Recapitulation of Parson^ s Method. I. Sampling, pulverization, and preservation of an air-dry portion in constant condition for an analysis. II. Estimation of moisture by loss at 100-120 C. III. Estimation of Ash. a. Portion of the ash soluble in water. b. Insoluble in water; soluble in dilute hydrochloric acid. c. Insoluble in water or in the acid ; soluble in sodium hydrate solution. IV". Estimation of the total nitrogen. For check on results ; for calculation of albuminoids after estimation of alkaloids, etc. Y. ESTIMATION OF PORTION SOLUBLE IN BENZENE (OB CHLORO- FORM). 1. Portion of benzene extract vaporized with benzene : volatile oils (camphors). 2. Portion of benzene extract soluble in water : alka- loids, glucosides, organic acids. o j Not soluble in water, | A. Ikaloids, possibly glu- ' \ soluble in dilute acid : j cosides. Removed bv animal Not soluble in water ) ! "' charcoal: "cMoro- pliyii, sortie resins. Not removed by the charcoal : some 4. < or the acid. Soluble > -j ( in alcohol of 80$. ) ft resins. ( Not soluble in water 5. < or the acid, or in al- V Waxes, fats, fixed ( cohol of 80$ : ) VI. ESTIMATION OF PORTION SOLUBLE IN ALCOHOL OF 80$ (after removal of Y.) The solution is made up to a defi- nite volume (200 c.c.) IN TWO EQUAL ALIQUOT PARTS (20 c.c. each) residues are obtained to furnish (1) the amount of organic matter, (2) the amount of organic matter soluble in water, thence the amount of or- ganic matter insoluble in water. IN " THE FIRST WAY " the constituents of YL, in the re- maining 160 c.c., are primarily divided according to their solubility in absolute alcohol, then by further treatment, as follows : PARSOXS'S METHOD. 419 A. Soluble in absolute alcoTiol. a. Soluble in water. Weight obtained. a' '. Precipitated by subacetate of lead. Tannin and most organic acids; some ex- tractives; some inorganic acids of the ash. Weight of all obtained. a" . Not precipitated by subacetate of lead. Alkaloids, glucosides, extractives, colors, a a f =. a" . b. Insoluble in water. Weight obtained. V . Soluble in dilute hydrochloric acid. Alkaloids, rarely glucosides, extractives. _"=:&'. ^ ~b" . Insoluble in dilute hydrochloric. Weight taken. b" 1 '. Soluble in dilute ammonium hydrate. Most acid resins, some colors. I" I"" - I'". V" . Insoluble in the ammonia. Weight taken. Neutral resins, some colors, albuminoids. B. Insoluble in absolute alcohol. c. Soluble in water. c'. Precipitated by subacetate of lead. Colors, extractives, rarely albuminoids, organic and inorganic acids. Weigh. c" . Not precipitated by subacetate of lead. Alkaloids, glucosides, glucose, sucrose, extrac- tives. c c' = c" . Estimate sugars. d. Insoluble in water. d f . Soluble in dilute hydrochloric acid. ' Alkaloids, glucosides. d d" d'. d" . Insoluble in dilute hydrochloric acid. Few resins, extractives, colors. Weigh. IN "THE SECOND WAY" the constituents of VI., taken in the remaining 160 c.c. of 80$ alcohol solution, are pri- marily divided according to their solubility in water, then by other treatment, as follows : (3) Evaporate to dryness, add water, reserve the residue (10), and make the filtrate up to 160 c.c. (4) In 20 c.c. estimate tannin. (5) In 20 c.c. estimate total precipitate by lead normal acetate. Tannins, acids (inorganic and organic), colors, extractives. 420 PL A NT ANAL YSIS. (6) In 20 c.c. estimate total precipitate by lead basic acetate. Compare precipitate with that in (5). In filtrate estimate the total glucose of sugars and glucosides. (7) In 20 c.c. duplicate the precipitation of (6). In filtrate estimate the actual glucose. Compare with total glucose. (8) In 20 c.c. triplicate the precipitation of (6). Ex- amine the precipitate for alkaloids, glucosides, sugars, extractives. (9) Use the remaining 40 c.c. for additional exami- nations. (10) The residue left in operation 3 may be tested for reams y albuminoids, colors, alkaloids, gluco- sides. VII. ESTIMATION OF THE PORTION SOLUBLE IN COLD WATER (after removal of V. and YI.) Examine as directed in the text, making up the filtrate to a definite volume, and taking aliquot parts (1, 2, 3, 4, etc.) for determina- tions and tests. In (1) determine total solids, and then the ash, to find the total organic substances. Gums, pectous substances, salts of organic acids, dextrins, soluble starches, albumens, colors. Examine by solubilities, iodine test, estimation of nitrogen, etc. The residue from solution VII., dried at 110 C., is weighed. VIII. ESTIMATION OF PORTION SOLUBLE IN BOILING- DILUTE ACID (after removal of Y, VI., and VII.) The weight of the washed residue obtained for estimation of the total solids of VIII. Starches estimated by determination of glucose with Fehling's solution, first examining for interfering extractives of a reduc- ing power. An aliquot portion of the liquid, freed from the sulphuric acid, is tested in portions quali- tatively. Small amounts of albuminoids may be found. IX. ESTIMATION OF PORTION SOLUBLE IN ALKALI-- WATER (after removal of portions Y. to VIII.) Take weight of insoluble washed residue, for estimation of total sol- ids. Album,ens, forms of pectin, humus, decomposi- tion products, colors. X. ESTIMATION OF THE RESIDUE LEFT BY SOLVENTS Y. to IX. Cellulose, lignose, colors, ash. Estimate from sepa- ration by chlorinated soda solution. PARSONS 'S METHOD. 421 Remarks. l It is advisable to determine always, in addition to what has already been directed, the amounts extracted directly from the sample by water, ether, alcohol of various percentages, methyl alcohol, benzin, chloroform, carbon disulphide, etc. In each extract estimate total organic matter and ash, and determine qualitatively, and quantitatively when possible, its constituents, by treating with such solvents and reagents as are indicated. Each extract being com- posed of certain distinct substances, it is necessary to account for them in every case. The amounts present of some constituents may be found by subtracting the weight extracted by some one solvent from the weight extracted by some other. It will be seen that this is a method of limited applicability, which can only be applied in those cases where the difference between the solvent action of the two liquids is very sharply defined. Certain special methods for the esti- mation of single constituents may be used, care being taken that all interfering substances be first removed. The methods of preparation of known substances as given in HUSEMANN'S " Pflanzenstoffe," and to a considerable extent in " Watts's Dictionary," may serve as suggestions for work. Treatment with ben- zene, 80 per cent, alcohol, and water, removes from nearly all plants the con- stituents of greatest chemical and medicinal interest, but in analyses of grains, fodder, and food materials, those compounds extracted by dilute acids and alka- lies have great value. There are substances in plants, seemingly isomers of starch and cellulose, which have properties more or less resembling those of cellulose, and are changed by boiling with dilute acids to glucose. In absence of an established nomenclature it has seemed best to use the terms "starch isomers" or " amylaceous cellulose " for these substances, 9 while those consti- tuents, not albuminous, which are removed by dilute alkali have been termed "alkali extract." These substances have been investigated by various chemists, but no definite and authoritative nomenclature has yet been adopted. THOM- SEtf gives the name " holz-gummi," 3 wood-gum, to a white substance extracted from plants by dilute sodium hydrate, while FREMY regarded these various com- pounds as modifications of pectic acid, pectin, and "cellulose bodies." 4 Starch also may exist in some seeds (as of sweet corn) in a form soluble in water. 5 It will be seen that the field for investigation is limitless, and that there is great need for improved methods for proximate analysis. The analyst will find that a study of any common plant will require of him much more than unthinking, mechanical habits of manipulation, while every careful investi- gation will reveal to him some constituents deserving more full and accurate study. 1 By Henry B. Parsons. 3 U. S. Dept. of Agric. Report, 1878, p. 189. 3 Jonr. prnk. Chem., 19, 146. *Compt. rend.. 83, 1136; Jour. Chem. Soc.. 31, 229(1877). 6 U. S. Dept. of Agric. Report, 1878, pp. 153-155. 422 PLANT ANALYSIS. O O n3 rd !:::>:::::! g ::::::: :| I 1 2 CO CO CO -OIUOWMY ..." '. \ \ \ '. : : '.&*:, , liiisliiiiiiiiiiiiiiiijii if a % c > a v- SV. _ a.*""* "SOOWWooooaooSjocBwrfapflS O_C O gj .>..->. i-> . jv. -. e- <-... e- S*-, p 2 ( e.. fi^.v. ev.cv o-. c 'S S co co co to to co^ HI 3 i-^r Ir-^^-3 (/J^-4^ c/j oj & m j/-",-;^ "S*-'^;^ g*iAiiiiiSiisssiiisii5iisi lift o : r = - r : o DRA GENDORFF'S ME THOD. 423 OUTLINE OF DRAGENDORFF' s METHOD OF PLANT ANALYSIS/ For the systematic analysis 30 to 50 grams may usually be taken. From 2 to 5 grams are dried at 100-110 for total moisture ; and usually another portion, not above 30 C., for amount of loss. The material for the systematic analysis to be powdered, sampled, mixed, and very finely pulverized for sol- vents. Yery hard bodies are dried at 100 to 110 C. before pulverizing. Fatty bodies may be lirst treated with the petroleum benziri. lor ignition pulverize very line, and if need be, after partial ignition, pulverize again. To promote combustion am- monium nitrate may be added, or ignited and weighed ferric ox- ide may be introduced. Powdered glass or washed sand may be intermixed. The carbon dioxide of the ash is to be determined. Special methods are used for the full quantitative estimations of distinct substances. I. SOLUTION BY PETROLEUM BENZIN (petroleum-ether, petrole- um spirit). This solvent to boil to the last at 45 G. and leave no residue. Use 10 c.c. for each gram of the dry plant pow- der. Macerate eight days, shaking daily. Aromatic fresh plants may be treated, without previous drying, by fine division, and by percolation with the solvent. Receive in a graduated separator, and take off aliquot volumes for examination and for weight of total dissolved substances. To evaporate the solvent from essential oils and other vola- tile matters, almost without waste of the latter, place 2 the solu- tion in a small, shallow dish, which is to be set within a wide- mouthed jar, this being so connected that a current of dried air is drawn over the surface of the solution. The air is drawn through chloride of calcium tubes, one of which is placed before and one beyond the jar containing the solution, so that there can. be no backward diffusion of moist air. The jar is closed air- tight by a stopper admitting entrance and discharge tubes, the entrance tube reaching nearly to the surface of the benzin solu- tion. The air is drawn at the desired rate by an aspirator, one acting by the discharge of water from a large closed bottle or jar. ' References to publication of DragendorfF s work are given on p. 407. This outline is by no means a substitute for Dr. DragendorfF s book on the chemistry and analysis of plants. But the outline of his plan of separations is presented for the convenience of a compact form, and as suggestion for instituting various analytical operations on vegetable tissues. * This method of evaporation in a current of dry air was used by OSSE, who reports control-analyses by it, 1876: Archiv d. Phar. [3] 7, 104; Jour. Chem. Soc., 29, 759; Dragendorffs " Plant Anal.," by Greenish, p. 21. 424 PLANT ANALYSIS. Fats may be treated with alcohol and observed with the microscope. 1 Glycerides may be saponified for separation [see pp. 274, 265, etc.] Alkaloids subjected to general tests [pp. 33, 42, 53]. Ethereal oils tested by solubilities, sensible properties, re- actions. Volatile acids recognized by acidity or by forming salts. Chlorophyll, by optical examination. 2 II. SOLUTION BY ETHEK. This solvent to be prepared as nearly as possible free from alcohol and from water (so as not to take up tannin). It is applied to the drug or vegetable matter previously exhausted by petroleum benzin, washed with the lat- ter, and dried. For 1 gram use 5 to 10 c.c. of the ether. Ma- cerate in a graduated cylinder seven or eight days. Take off aliquot volumes for examination. Evaporate the ether by a current of dried air, as directed for the benzin. Test portions by solubilities in (a) water, (b) alcohol (absolute), (c) alkali. The ether-soluble portion may contain benzoic, salicylic, and gal- lic acids, salicin and other glucosides, alkaloids, resins, hema- toxylin, etc. For estimation of total ether-soluble fixed sub- stances an aliquot part is evaporated and dried at 110 C. [Fur- ther see the articles Alkaloids, Benzoie Acid, etc. Resins are found in the part insoluble in water. Compare p. 278.] III. SOLUTION BY ABSOLUTE ALCOHOL (following solvents I., II.) For 1 gram of the material 10 c.c. of the sol vent. Macerate five to seven days, restoring loss, then filtering through paper wet with alcohol. Evaporate an aliquot volume, and dry at 110 C. for weight. Evaporate other portions, without heat, in vacuum, and dry over sulphuric acid. A residue, obtained as last directed, is to be treated with water, in measured proportion, filtered, the filtrate evaporated to constant weight at 110 C., for weight of all water-soluble matters in III. Other portions of the aqueous solution are taken for the estimation of tannins and for sugars. A portion of the residue (III.), undissolved by water, is gently dried and treated with ammonia water dilute (1 : 50), the ammonia solution acidulated with acetic acid, and the precipitate, if any,, after concentration, examined for phlobaphene, which may be estimated in this way [see Phlobaphene, under Tannins]. Por- tion III. may contain resins, alkaloids, glucosides, bitter prin- ce 1 See HEINTZ: Ann. Ptiys. Chem. (Pogg.), 92, 588; Phar. Jour. Trans. [1] 15, 425. GREENISH: Phar. Jour. Trans. [8] 10, 909. This work, p. 297. 2 Dragendorff, English ed., p. 19. '-'"N/VERSITY DRAGENDORFFS METHOD. 42$ The water-soluble portions of II. (the ether -extract) and of III. (the alcohol-extract) may be treated by imm-iscible solvents^ applied first to the watery liquids made acidulous with sulphuric acid, and then applied to the same liquids made ammoniacal with ammonia. [See this work, pages 33 and after ; and the author's " Outlines of Proximate Organic Analysis," p. 136.] As immis- cible solvents, petroleum benziu, benzene (boiling constant at 81 C.), and chloroform are recommended. The following re- sults are indicated : By petroleum benzin from acid solution Absinthin, Capsicum, Hop bitter, Piperin, Salicylic acid. By benzene from acid solution Absinthin, Berberine, Caffeine, Caryophyllin, Cascarillin, Colchicine, Colocynthiri, Cubebin, Daphnin, Elaterin, Ericolin, Gratiolin, Menyanthin, Populin r Quassin, Santonin. By chloroform from acid solution JEs- culin, Benzoic acid, Cinchonine, Colchicine, Convallamarin, Di- gitalein, Helleborin, Narceine, Physalin, Picrotoxin, Quinidine, Theobromine, Saponin, Senegin, Solanidin, Syringin. (Before making alkaline, the dissolved chloroform is washed out with a little petroleum benzin.) By petroleum benzin ^rom alkaline aqueous solution Brucine, Capsicum, Conine, Emetine, Lobe- line, Morphine, Nicotine, Sabadilline, Sabatrine, Sparteine, Strychnine (traces), Trirnethylarnme. By benzene from alka- line solution Aconitine, Atropine, Cinchonme (traces), Co- deine, Delphinine, Gelsemine, Hyoscyamine, Physostigmine, Pilocarpine, Nareotine, Quinidine, Taxine. By chloroform from alkaline solution Cinchonine, Morphine (traces), Papa- verine, Narceine. By amyl alcohol from alkaline aqueous so- lution (following previous solvents) Morphine, Salicin, Sola- nine. [Tests for a glncoside, by fermentation, are indicated in the article Tannins in this work. Estimation of alkaloids, p. 44, and under the several alkaloids.] IY. SOLUTION IN WATEK. The residue insoluble in absolute alcohol (III.) is dried and treated with 10 parts of water, by 48 hours' digestion, then filtered through the filter previously used. The filter should be washed with water, and the washings exam- ined separately. An aliquot volume (10 to 20 c.c.) of the fil- trate is evaporated, and dried at 110 C., for weight of total sub- stances in IV. To another aliquot portion, 10 to 20 c.c., add 2 c.c. absolute alcohol, leave 24 hours in a cool place, filter on a tared filter, wash with 66$ alcohol, dry, and weigh. Find the weight of ash in each of the two portions last weighed. In IV. may be found pectous substances, albumens, inulin, dextrines, sugars, acids, saponin. A precipitate by lead acetate will 426 PTOMAINES. contain the acids, with mineral acids. Sugars here are to be estimated. A portion, before and after obtaining IV., may be subjected to estimation of nitrogen, when consideration is given to the presence of amrnoniacal salts, amides, alkaloids, nitrates, etc. (Dragendorff, paragraph 97). Albumen is estimated by pre- cipitation with tannin or from the amount of nitrogen. Y. SOLUTION IN ALKALI WATER. For 1 part residue not dis- solved in IY. take 10 parts of a 0.1 to 0.2$ solution of sodium hydrate. Macerate 24 hours. Filter an aliquot volume, satu- rate with acetic acid, add alcohol of 90$, leave 24 hours in the cold. Collect the precipitate on a tared iilter, wash with 75$ alcohol, dry, and weigh. Ignite and weigh the ash. Al- bumens and pectous substances are contained. The residue insoluble in alkali is apt still to retain traces of nitrogen com- pounds. YI. SOLUTION IN ACIDULATED WATER (after removal of I. to Y.) The residue not soluble in Y., washed, is treated with a 1$ solution of hydrochloric acid. It is found by a microscopic examination of the original material, whether it contains starch or not. Oxalate of calcium may be separated and estimated, digesting with the acid for 24 hours at 30 C. In a measured quantity of the liquid, neutralize with ammonia, add acetate of sodium to react with all the hydrochloric acid, and set aside for the calcium oxalate to form, for gravimetric determination (as calcium carbonate). VII. THE INSOLUBLE RESIDUE from VI. is washed, dried, and weighed. Treated with chlorine, and weight of residue found, the difference represents lignin and incrusting substances ; the remainder contains the cellulose, which is examined microsco- pically, and the ash. PROTOPINE. See OPIUM ALKALOIDS, p. 360. PSEUDACONITINE. See ACONITE ALKALOIDS, p. 19. PSEUDOMORPHINE. See p. 359. PTOMAINES. Cadaveric Alkaloids. Ptomaine. Alka- loid-like bodies formed in the putrefactive decomposition of animal tissues. The formation may commence shortly after death. It may be caused or promoted by digestion with acids (COPPOLA, 1885), especially when the mixtures are acidulated with sulphuric acid. BRIEGER (1885) enumerates the following products of the putrefaction of the human body : PTOMAINES. 427 Choline, C 5 H 15 NOo. Eeuridine, CgjL^No. Cadaverine, C 5 H 16 N 2 , boiling at 115-120 C., with water at 100 C. Putrescine, C 4 H 12 N 2 , boiling at about 135 C. Saprine, C 5 H 16 IS 2 . Trimethylamine, C 3 H 9 N. Mydalein. A ptomaine boiling at 284 C. Of the above, only Choline is poisonous. From putrefactive albumen and gelatine Brieger (1885) had obtained : Neurine, C 5 H 13 NO. ../ Muscarine, C 5 H 15 NO 3 . An ethylenediamine, C 2 H 4 (H 2 N) 2 . Neuridine, C 5 H 14 K 2 . Gadinine, C 7 H 17 JtfO 2 . Triethylamine, dimethylamine, and trimethylamine. Of these the first three named are extremely poisonous. The greater number of cadaveric ptomaines are non-poisonous. Concerning their chemical constitution, Brieger (1885) calls attention to the fact that most of the characteristic ptomaines are diamines ; that they are chemically more simple in composition than are the vegetable alkaloids ; that many of the ptomaines are derivatives of hydrocarbons of the ethylene series, and are in distinction from true alkaloids representing the pyridine group. Ptomaines are for the most part obtained from tissues in the operations of separation by the immiscible solvents. In evapo- rations they are liable, in part, to be vaporized. They are easily decomposed and are affected by atmospheric oxidation. They respond to the greater number of the general reagents for preci- pitation of alkaloids. Cadaverine hydrochloride, C 5 H 16 N 2 .2HC], gives reactions as follows (BRIEGER, 1885): With phosphomolybdic acid, a white crystalline precipitate. With iodide of potassium, or with iodine in iodide solution, brown needles ; with potassium bismuth iodide, reddish needles ; with picric acid, yellow needles ; with potas- sium chromate and concentrated sulphuric acid, a red-brown pre- cipitate, soon vanishing. Free cadaverine gives with potassium mercuric iodide a resinous precipitate ; with potassium iodide, a brown precipitate ; with tannic acid, a white precipitate. Free or in salt it promptly reduces a mixture of ferric chloride and potassium ferricyanide, giving a blue color. 428 PTOMAINES. Putrescine hydrocliloride, C 4 H 12 N 2 . 2HC1, with phosphomo- lybdic acid gives a yellow precipitate ; with potassium mercuric iodide, an amorphous precipitate soon crystallizing in needles ; with iodide of potassium, or iodine in iodide solution, a brown crystalline precipitate. Ptomaines are mostly quite strong reducing agents, and the reaction of ptomaine sulphates with ferric chloride and potas- sium ferricyanide (BROUARDEL and BOUTMY, 1881) has been an- nounced as characteristic of the animal alkaloids, but this is not admitted by Brieger. The latter states that the reduction of the ferricyanide mixture, with formation of a blue color, is obtained by cadaverine, saprine, mydaleine, and some other ptomaines, not by choline, neuridine, or putrescine. Brieger further states that he has not found a distinctive reaction for ptomaines. They are all precipitated by phosphomolybdic acid, a reaction they share with ammonia, as well as with the vegetable alkaloids (p. 46). The reduction of ferricyanide with ferric salt, forming prussian blue, is not given promptly by many vegetable alka- loids, but is given at once by morphine and veratrine (Brouardel and Boutmy), at once by colchicine (Beckurts, 1882), and is given slowly and feebly by aconitine, brucine, conine, digitaline, nicotine, strychnine, papaverine, narceine, codeine (Beckurts). LETJCOMAINES. Animal alkaloids, more or less septic, formed in tissues and organs of the living body : Xanthocreatinine, C 5 II 10 N 4 O. From muscular tissue. Ee- semblee creatinine. Cruscocreatinine, C 5 H 8 N 4 O . . Kesembles creatinine. Arnphicreatinine, C 9 H 19 jN 7 O 4 . Pseudoxanthine, C 4 H 5 N 5 O . . . Eesembles xanthine. Mytilotoxine, C 6 H 15 NO 2 From mussels ; poisonous. Betaine, C 5 H 1:L NO2 From mussels; non-poisonous. The first four above given were found by GAUTIER (1886) ; the last two by BRIEGER (1886). Neurine was obtained by MARINO-ZUCO (1885) from fresh eggs, blood, brains, liver, etc., by the method of Stas, and more abundantly by the method of Dragendorff, and formed from the lecithin of the tissues by action of acids on them, not formed from the albuminoids. These leucomaines mask the reactions for the vegetable alkaloids. By repeatedly extracting (shaking out) from alkaline solution, with ether or chloroform, it was found that the neurine was left behind. From the liver and spleen, in addition to neurine, a violet fluorescent base was obtained. PTOMAINES. 429 Respecting cheese-poison, reported upon by VICTOR C. YAUGHAN in 1884 as a poisonous ptomaine, and now announced by him as diazobenzene salts, see Tyrotoxicon, in this work. The literature of ptomaines and leucomaines is mainly embraced in that of physiological and pathological chemistry. Among the publications of interest in analytical chemistry and toxicology, an index is here made of the following : DUPRE and BENCE JONES, 1866 : Respecting a frail alkaloid-like body found in the organs and liquids of the bodies of man and of animals, Zeitsch. Chem. und Phar., 1866 ; Phar. Centralh., 16 ; Ber. d. chem. Ges., 7, 1491. SONNENSCHEIN and ZULZER, 1869 : On bases obtained from mus- cular tissue, Berlin Jdin. Wochenschr., 1869, 123. RORSCH and FASSBANDER, 1871 : On a body giving reactions for alkaloids, found in analyses of liver, etc., for poisons, Ber. d. chem. Ges., 7, 1064. SSLMI, chiefly about 1878 : On toxicology, 1876, Gazzetta thim. ital., 4, 1 ; Jour. Chem. Soc., 27, 607. On alkaloids of cada- veric putrefactions, 1873 to 1880: Ber. d. chem. Ges.,^ 142; 8, 1198; 9, 195; II, 808, 1838; 12, 279; 13, 206. " Sulle Ptomaine ad alkaloidi cadaverici," Bologna, 1878. On alkaloids in the cadaver, 1879, Gazzetta chim. ital., 9, 35 ; Jour. Chem. Soc., 36, 734. On a poisonous alkaloid from a cadaver containing arsenic, 1 879 : Gazzetta chim. ital., 9, 33; Jour. Chem. Soc., 36, 734. On an alkaloid found in the brain and liver, and in the wild poppy, 1876, Gazzetta chim. ital., 5, 398 ; Jour. Chem. Soc., 29, 938. On pathological bases, 1881, Gazzetta chim. ital., 1881, 546; Jour. Chem. Soc., 42, 741. TH. HUSEMANN, 1881 : The ptomaines in toxicology, Archiv der Phar. [3] 16, 415 ; Am. Jour. Phar., 54, 152. H. BECKURTS, 1882 : Distinctions between cadaver and plant al- kaloids, Archiv der Phar. [3] 17, 104 ; Am. Jour. Phar., 54, 221. Zeitsch. anal. Chem., 24, 485. BROUARDEL and BOUTMY, 1881: Distinctive reactions of pto- maines, Ber. d. chem. Ges., 14, 1293; Compt. rend., 92, 1056. MARINO-ZUCO, 1884 : Ptomaines in toxicology, Gazzetta chim. ital., 13, 431, 441 ; Jour. Chem. Soc., 46, 342, 343. ARNOLD, 1884 : Ptomaines in toxicology, Archiv der Phar. [3] 21, 435 ; Jour. Chem. Soc., 46, 469. GARNIER, 1883 : Ptomaines in toxicology, Jour, de Phar., 7, 377 ; Am. Jour. Phar., 55, 404. L. BRIEGER, Berlin, 1884-87 : " Ueber Ptomaine," Berlin, 1885. 430 PYROGALLOL. Zeitsch. physiolog. Chem., 3, 135 ; 9, 1. " Weitere Unter- suchungen uber Ptomaine," Berlin, 1885-86. Ber. d. chem. Ges., 17, 274:1 ; Zeitsch. anal. Chem., 24, 4:84:. A new pto- maine producing tetanus, Ber. d. chem. Ges., 19, 3119 ; Jour. Chem. Soc., 52, 284. MAAS and others, 1884: Ptomaines in boiled meat, Chem. Cent., 1884, 975 ; Jour. Chem. Soc., 48, 676. Y. C. YAUGHAN, 1884-85 : A ptomaine from poisonous cheese, Zeitsch. physiolog. Chem., 10, 146; Jour. Chem. Soc., 50, 373. Michigan State Board of Health Reports. (See " Ty- rotoxicon," in this work.) COPPOLI, 1885 : Ptomaines formed by processes of analysis of tissues for poisons, Gazzetta chim. ital., 14, 124, 571; Jour. Chem. Soc., 48, 278, 913. GAUTIER, 1885-86 : Leucornaines, Bull. Soc. Chim., 43, 158 ; Jour. Chem. Soc., 48, 676. On alkaloids of bacterial origin, 'etc., Paris, 1886. Ptomaines and Leucomaines, 1886, Jour. Phar. [5] 13, 354; Jour. Chem. Soc., 50, 634. LADENBURG, 1885: Ber. d. chem. Ges., 18, 2956, 3100. OLIVERI, 18&6 : Supposed ptomaines of cholera, Gazzetta chim* ital., 16, 256 ; Jour. Chem. Soc., 50, 1049. PURPURINE. See COLORING MATERIALS, p. 190. PURPUROGALLIN. See p. 431. PYROGALLOL. C 6 H 6 O 3 = C 6 H 3 (OH) 3 = 126. Pyro- gallic Acid. Manufactured from gallic acid or from gallotan- nin by sublimation. One part of gallic acid with two parts of powdered pumice stone may be heated to 210 -220C. in a stream of carbon dioxide. To obtain colorless, sublimed in a vacuum at 210 C. Used as a reducing agent in photography ; also to a limited extent in hair dyes, either by itself or to reduce silver. Pyrogallol is identified by its reactions with alkalies and iron salts, and its formation of purpurogallin. It is separated from tannic acid by its not precipitating with gelatin. It may be estimated in a lead compound. Pyrogallol crystallizes in lustrous plates or needles of white or yellowish- white color, a very bitter taste, without odor, and a neutral or very feebly acidulous reaction. It gives a brown color to the skin. The crystals are changeless in dry, pure air, dark- ening in ammoniacal air. It melts at 115 C., boils at 210 C., PYROGALLOL. 431 and at about 250 C. blackens with production of metagallic acid. (See Gallic Acid, p. 321.) It dissolves in three parts of water, freely in alcohol and in ether, not in absolute chloroform. The watery solution darkens on standing, sooner if heated, quickly coloring by addition of alkalies, with formation of alkali carbonate and acetate, absorption of oxygen taking place to an extent pro- portional to the coloration, which is destroyed by oxalic acid. The alkalies cause reddish-yellow to red -brown tints ; lirne solu- tion, a violet to purple color ; all becoming gradually brown to black. Ferroso ferric salts, slightly oxidized ferrous salts the better, give a clear blue color. If there be much ferric salt the color soon turns to red, and with ferric salt alone the color is reddish at first. If the very dilute solution of ferric salt and pyrogallol be gradually treated with ammonia, the color changes first from red to blue, and then back to bright red. (The reaction is like that of purpurogallin, given below.) By gradually adding then acetic acid or other organic acid, the blue is first restored, then a red color again appears. Hydrochloric acid and most inorganic acids give at once a red color. The blue color is produced by bicarbonates as well as by ammonia, 1 also by free alkaloids (SCHLAGDENHAUFFEN). In presence of gum arabic, blood, saliva, and various other organic substances, pyrogallol, in solution, exposed to the air, gradually forms PURPUROGALLIN, C 20 H 16 O 9 (STRUVE). The same product is obtained at once by adding a strong solution of permanganate acidulated with sulphuric acid. Purpurogallin has a red color of much intensity, imparted to solutions, from which it crystallizes in yellow to red needles, and by sublimation is obtained in garnet-red crystals. It is sparingly soluble in water, and its solution, with an alkali, gives a transient blue color of great intensity. 2 Pyrogallol is a most forcible reducing agent, promptly re- ducing salts of silver and mercury, and Fehling's solution, and reducing ferric salts in the iron reactions above given. It is 'JACQUEMIN, 1874 and 1876-77: Ann. Chim. Phys. [4] 30, 566; Jour. Client,. Soc., 27, 1016; Jour. Chem. Soc., 31, 340. A very dilute solution of ferric chloride and pyrogallol is used as an indicator, more delicate than litmus, for the estimation of ammonia or of bicarbonates (as in mineral waters). The solution is made of equal volumes of a solution of 5 grams pyrogallol to the liter, and a solution of 2 grains ferric chloride to the liter. It deposits purpu- rogallin, and needs to be filtered from time to time. Of the solution 10 c.c. are added to 250 c.c. of water for alkalimetry. 4 As to ethers of pyrogallol, and their color products, see " Watts's Diet.," viii. 1710. Pyrogalloquinone, ibid. 1713. Reaction with mercuric chloride and alkaloids, ibid. 1709. 432 RACE MIC ACID. attacked by nitric acid, with red products. Its dry mixtures with many oxidizing agents are explosive. In aqueous solution, with alkali, it removes nearly all the oxygen from a contined portion of air. A gravimetric estimation may be made by adding to an alcoholic solution of pyrogallol an alcoholic solution of lead acetate, faintly acidulated with acetic acid, quickly washing the precipitate with alcohol, drying on a water-bath, and weighing. rb(C 6 H 5 O 3 ) 2 : 2C 6 H 6 O 3 :: 457 : 252 :: 1 : 0.5514. QUINAMINE. See CINCHONA ALKALOIDS, p. 92. QUINICINE. See p. 94. QUINIDINE. See p. 154. QUININE. See p. 125. QUI.NOIDINE. See p. 94. QUINOLINE. See p. 165. QUINOLINE RED. See COLORING MATERIALS, p. 182. RACEMIC ACID. H 2 C 4 H 4 O e = 150. Paratartaric Acid. Traubensaure. Separable Inactive Tartaric Acid. An isomer of tartaric acid, found in some varieties of grapes, and differing from dextrotartaric acid in the form of crystallization, in optical powers, and in its solubilities as free acid and as calcium salt. It crystallizes in the triclinic system, with one molecule of water, becoming anhydrous at 100 C. It is soluble in about 5 parts of cold water and in 48 parts alcohol of 0.809 specific gravity. Its solution is optically inactive, not rotating the plane of polarized light, but it is separable into dextrotartaric and levotartaric acids, as follows : When the racemates of two bases, as sodium and ammonium, in molecular proportions, are crystallized from solu- tion together, crystals of a double salt, as NaNH 4 C 4 II 4 O 6 , are obtained, and these crystals, rectangular prisms, have certain hemihedral faces, and are divided into pairs, right and left, by the position of the hemihedral faces. The one crystal of a pair coincides with the reflection of the other from a mirror. When the two kinds of crystals are separated by hand-picking, the one kind is found to be the salt of dextrotartaric acid, identical with ordinary tartaric acid, while the other kind is a salt of another SALICYLIC ACID. 433 tartaric acid isomer, whose solution rotates the light plane to the left, and is termed Levotartaric Acid, or anti tartaric acid. Racemic acid, free, forms a precipitate with calcium sulphate solution on standing, and a precipitate with calcium chloride solution quite readily ; also, the calcium precipitate, dissolved by hydrochloric acid, is precipitated again by ammonia (distinc- tions from dextrotartaric acid). RHOEADINE. See OPIUM ALKALOIDS, p. 360. RICINOLEIC ACID. See FATS AND OILS, pp. 246, 248, 289. RESIN, COMMON, SEPARATION OF. See FATS AND OILS, p. 278. ROSIN OILS. See p. 280. SAFFLOWER RED. See COLORING MATERIALS, p. 191. SAFFRANINS. See p. 183. SALICYLIC ACID. Salicylsaure. Acide Salicylique. O 7 H 6 O 3 = 138 (monobasic and with alkalies feebly dibasic). In structure, C 6 H 4 .CO 2 H.OH, in which CO 2 H : OH = I : 2, or- tho-hydroxybenzoic acid. There is but one salicylic acid, but it is one of three isorneric hydroxybenzoic acids (or phenol- car- boxylic acids), namely, the ortho, meta, and para compounds, with the respective positions of 1:2, 1:3, and 1 : 4, for C0 2 H : OH. Sources. Free salicylic acid occurs very sparingly in nature, having been found in the flowers of Spiraea ulmaria, in Viola tricolor and other species of viola, and in the Gloriosa superba of the East Indies. The ethereal salt, salicylate of methyl, C 6 H 4 .CO 2 (CH 3 ).OH, is known as "wintergreen oil." Methyl salicylate forms the larger part (over 99 per cent., PETTIGREW, 1884; 90 per cent., CAHOURS, 1843) of the oil of gaultheria, IT. S. Ph. ; according to Pettigrew the whole of the " oil of birch," from Betula lenta bark, commonly sold as " wintergreen oil " ; and nearly or quite the whole of the oil of Andromeda Leschenaultii, of abundant growth in Hindostan, and the volatile, oil of Mono- tropa Hypopitys of northern Europe. It is also found in the oils of several species of Gaultheria and in oil of Poly gala pauci- flora ; sometimes in oil of cloves, and in oil from buchu leaves. 434 SALICYLIC ACID. Salicylic acid may be prepared from methyl salicylate by boiling with potassium hydrate solution until the oil is dissolved, and as long as methyl alcohol is given off, and then acidulating with hydrochloric acid, when the salicylic acid precipitates. Since 1874 salicylic acid has been extensively manufactured from carbolic acid by Kolbe's method. 1 Dry sodium phenol, C 6 H 5 ONa, is treated with dry CO 2 , at a temperature increased to 180 C. and finally to about 225 C., whereby disodium sali- cylate, C 6 H 4 .CO ? Na.O]S"a, is formed in the retort, and half the phenol taken is distilled over. Small portions of para-hy- droxybenzoic acid and traces of a phenol-dicarboxylic acid are formed (OsT, 18T9). If potash be used instead of soda the pro duct is the para-hydroxybenzoic acid. But impurities of greater proportion in salicylic acid made from -carbolic acid probably result from the impurities in the latter, namely, from the cresols homologous with phenol (the " cresylic acid " ) present in carbo- lic acid (see Phenol). Each of the three cresols, C 7 H 8 O, treated with sodium and carbon dioxide, forms a cresotic acid, C 8 H 8 O 3 . The cresotic acids so formed are sometimes termed the liomo- salicylic acids, and are direct homologues of salicylic acid, hav- ing the rational formula, C 6 H 3 (CH 3 ) . CO 2 H . OH, with the posi- tions CO 2 H : OH : CH 3 = respectively 1 : 2 : 3, and 1 : 2 : 4, and 1:2: 5. 2 1 Concerning recent manufacture of salicylic acid through formation of diphenyl carbonate, at the works of Aktien (late Schering) in Berlin, see Jahr. chem. 'Tech., 1884, 504; HENTSCHELL, Jour, prakt. Chem., 27, 159, and Jour. Soc. Chem. Ltd., 3, 115, 646. 2 These three cresotic acids are but three isomers among ten known isome- ric hydroxytoluic acids (or cresol-carboxylic acids) obtained from various sources. Beilxteiiis Organisch. Chemie. 1883, p. 1457. In part in " Watts's Diet.," viii. 2'23. Concerning certain of these acids and xylene products, Am. Chem. Jour.. 3, 424; 4, 186. Concerning three hydroxyxylenic acids, GUNTER, 1884: Ber. d. chem. Ges., 17, 1608; Jour. Chem. Soc.', 1884, Abs.. 1347. It will be observed that the occurrence of homologues in salicylic acid from coal- tar corresponds to the existence of homologues in carbolic acid and in the benzoles, as follows: 1. Benzene C 8 H 6 . Toluene C 7 H 8 . Xylenes C 8 H] , 2. Phenol C 6 H 6 0. Cresols (see p. 394) C 7 H 8 0. Xylenols C 8 H 10 . 3. Benzoic acid C 7 H 6 2 . Toluic acids C 8 H 8 2 . Xylenic acids CgHj 2 . 4. Salicylic and two other hydroxybenzoic acids . C 7 H 6 3 . Cresotic and other hy- droxytoluic acids C 8 H 8 3 . Hydroxyxylenic acids. . . C 9 Hi 3 . SALICYLIC ACID. 435 The question of the occurrence of the homologues and iso- mers of true salicylic acid, in the article made from carbolic acid, is further treated under Impurities (g). At all events, the crude sodium salicylate of Kolbe's process is acidulated with hy- drochloric acid, and the resulting crude salicylic acid is purified in various ways, by crystallizations from dilute alcohol or hot water, by dialysis of the sodium salt, and by filtration through purified animal charcoal. Dr. Squibb (1877) employed sublima- tion of the acid by heat from steam. For some years the " natu- ral salicylic acid " has been manufactured from " wintergreen oil" in this country, for medicinal purposes, with claims for supe- rior purity. The essential oil of the flowers of Spiraea ulmaria, Salicylol) is the aldehyd of salicylic acid. The glucoside Sali- cin, the active principle of Salix, readily liberates the correspond- ing alcohol, saligenin. From all these substances, from indigo, and from coumaric acid, salicylic acid can be obtained by heat- ing with potassium hydrate under suitable conditions, and by other chemical treatment. SALICYLIC ACID is identified by its crystalline form and physi- cal deportment (#), its reaction with ferric salt and with nitric acid, and the odor of its methyl ester (d). From Benzoic acid it is distinguished by the odor of the respective products with so- dium amalgam in presence of water, and with lime by heating when dry (d) ; from Cinnamic acid by the permanganate oxida- tion of the latter to benzoic aldehyde. It can be separated and its valuation secured (e) by distillation from its salts (1), or of the free acid (3) ; by solvents not miscible with water (2) ; by dialy- sis (4) ; and in special methods from Wine and Beer (p. 440), Canned Fruits, Milk (p. 440), and the Urine (p. 441). Quantita- tively it is estimated (f) by the colorometric method, or weighed as free acid. It is examined respecting impurities and require- ments of quality (g) with regard to its modes of manufacture (p. 434) and its chemical isomers (p. 443), by application of re- cognized special tests (p. 444). For Salicyluric Acid see p. 445; Salicylate of Sodium, p. 445 ; other salicylates, p. 437. a. Salicylic acid is furnished, according to its grade, in fine, needle-shaped crystals, or in a loose or granular powder, obscure- ly crystalline or nearly amorphous. White when pure, it is fre- quently blemished with a yellowish, or pinkish, or brownish tinge. The dialyzed acid is said to keep perfectly white. The " recrystallized " acid is a good pharmacopoeial brand ; the " pre- cipitated " acid is of a lower grade ; the " sublimed '' acid is said 436 SALICYLIC ACID. to acquire color and carbolic odor. The crystals are monoclinic (MARIGNAC, 1855). From moderately warm aqueous solution it is obtained in four-sided prisms, from hot aqueous solution in needles, from alcoholic solution by spontaneous evaporation in large four-sided columns, from a drop of ether-solution evapo- rated on a glass slide in star form or feathery groups of radiate needles requiring to be magnified 50 to 100 diameters (HAGEE). Sp. gr. 1.483 at medium temp., taking water at i C. as 1 (SCHROEDER, 1879). Permanent in the air. Melts at 156 C. (312 F.) (HiiBNER, 1872 ; KOHLEK, 1879). Sublimes unaltered by heat from steam at 60 to 80 Ibs. pressure, not above 145 C. (293 F.), the product having no carbolic odor (SQUIBB, 1883). Suddenly heated, at 220-230 C. (428-446 F.), it is resolved into phenol and carbon dioxide, leaving no residue, and when sublimed without care the sublimate is contaminated with phe- nol and gives a carbolic odor In boiling its aqueous solution it vaporizes unaltered with the steam. Heated with concentrated hydrochloric or dilute sulphuric acid, under pressure, at 140- 150 C., it is dissociated into phenol and carbon dioxide. Alkali salicylates oxidize readily. J. Pure salicylic acid is odorless and has a sweetish, acidu- lous, acrid taste. The acid of commerce sometimes has the odor of phenol or of cinnamic acid. Salicylic acid is not caustic, but is somewhat irritant to mucous surfaces, the more so by inhala- tion in dust. It is medicinal in ordinary doses of 10 to 60 grains. If more than 150 grains be given within twe,nty-four hours some disturbance usually follows. The alkali salicylates have the effect of the acid, as does also methyl salicylate (Gaultheria oil) (H. C. WOOD and H., 1886). In hypodermic injections about 0.2 per cent., or the strength of a cold saturated aqueous solution of the acid, is employed. For external application 1 to 10 per cent, solutions are used. Salicylic acid is removed from the system with moderate rapidity, most largely by the urine, in part by the bile, and in traces by the saliva. Gaultheria oil becomes free salicylic acid in the living body (Wooo, 1886 : Ther. Ga- zette, 10, 73). In the urine salicylic acid is excreted as salicyluric acid, with unchanged salicylic acid. Other reported excretory products are phenol, salicin, and indican. Salicylic acid, in proportion of about 0.1 per cent. (} grain to the fluid- ounce), preserves ordinary vegetable infusions. For fruit juices, cider, etc., 0.05 to 0.3 per cent, is requisite, but 0.01 to 0.02 per cent, exerts a degree of conserving power. To preserve hypodermic and alkaloidal solutions Dr. Squibb uses the full or half strength SALICYLIC ACID. 437 of a cold water saturated solution (0.2 or 0.3$). * As an antizymotic, or antiseptic, the salicylates have much less power than the free acid, and a sufficient quantity of bisulphate of potassium may be added to complex liquids, with salicylic acid, to prevent its com- bination with the bases of acetates, etc. The use of salicylic acid in foods has been forbidden in some countries. G. Salicylic acid is sparingly soluble in water: at 15 C. (59 F.) it requires 444 parts ; at 20 C. (68 F.). 370 parts ; at 30 C. (86 F.), 256 parts; at 100 C. (212 F.), 13 parts (Boun- GOIN, 1879). Heated with water under pressure, the acid dis- solves water and liquelies (ALEXEJEFF, 1883). In alcohol of 90$ it dissolves in 2.4 parts at 15 C. (59 F.) ; in absolute alcohol, in 2 parts at 15 C. It dissolves, at 15 C. (59 F. ), in 2 parts of ether, in 3.5 parts of amyl alcohol, freely in methyl alcohol, in 80 parts of benzene, in 80 parts of chloroform, in 60 parts of glycerin, and in about 60 parts of ordinary fixed oils. It dissolves in carbon disulphide and in volatile oils. 2 Salicylic acid has an acid reaction ; it causes effervescence from carbonates ; it forms moderately stable monobasic or normal salts (as CgE^.COjjNa. OH), those of the alkali metals being neutral to litmus (when pure), and instable dibasic salts (as CgH^.COgNa.OJSra) of alkaline reaction. For conserving certain alkaloids the salicylate is a very favorable salt. Of the normal salts, those of alkalies, calcium, barium, magnesium, zinc, and copper dissolve in water, the lead salt in hot water, but the silver salt does not readily dissolve. The basic salts of non- alkali metals are not soluble in water. Salicylate of quinine, C2 H 24 ]Sr 2 O 2 .C 7 H 6 O34(H 2 O), is neutral and soluble in 900 parts of cold water or 20 parts of alcohol ; salicylate of atropine is neutral and very soluble in water. With the alkali salts of the weaker acids salicylic acid dissolves freely in water, making comparatively concentrated solutions, which, however, are really solutions of salicylates. Borax, acetate of potassium, and acetate of ammonium are used, also alkali phosphates and citrates, with water, as solvents. With borax a crystallizable union is obtained, NaC 7 H 5 O 3 + C 7 H 5 (BO)O 3 , of acid reaction. With half its 'ROBINET and PELLET, 1882: Compt. rend., 94, 1322; Jour. Chem. Soc., 42, 1010. BERSCH, 1882: " Biedermann's Centralblatt," p. 340; Jour. Chem. Soc., 42, 1010. HEINZELMANN, 1884: "Biedermann's Centralblatt," p. 503; Jour. Chem. Soc., 46, 764. " Hager's Phar. Praxis," Erganzungsband, 43. SQUIBB'S Ephcmeris, 1882-85: i, 414; 2, 833. 2 LANGBECK (1884) reports widely varying solubilities of salicylic acid in different essential oils, and uses this difference to distinguish volatile oils from each other: Repert. anal. Chem., 12, 177; Jour. Soc. Chem. Ind., 3, 547. 438 SALICYLIC ACID. weight of borax, and 2J times its weight of glycerin, a 25 per cent, solution of salicylic acid may be obtained. The combina- tion with boric acid, borosalicylic acid, C 7 H 5 (BO)O 3 .C 7 H 6 O 3 , is soluble in 200 parts of cold water (HAGER). Glycerin sali- cylate can be formed (GorriG, 1877). Solutions of salicylic acid and its salts are not easily preserved, and acquire color by standing. d The stronger acids precipitate salicylic acid from solu- tions of its salts in less than 200 to 400 parts of water. Silver nitrate solution, with solutions of saiicylates, not with solution of salicylic acid, forms a white precipitate of silver salicylate, C 7 H 5 AgO 3 , dissolved by boiling water, also by nitric acid and alcohol. Ferric chloride solution, with solutions of salicylic acid or its salts, gives (according to dilution) a violet-blue to violet-red color of great intensity. The alcoholic solution of free acid is most favorable. A little less delicate than the sul- phocyanide reaction (E. F. SMITH, 1880), it reveals salicylic acid diluted to 100000 parts (ALMEN, 1878). The reaction is pre- vented by alkalies, and hindered by alkali acetates, phosphates, borates, potassium iodide, and by oxalic, tartaric, citric, phos- phoric, and arsenic acids, not by dilute acetic, boracic, sulphuric, or nitric acid, nor by glycerin, alcohol, ether, common salt, or nitre (HAGER, 1880). The ferric violet reaction is given by sali- cyluric acid and oil of spiraea, not by para or by rrieta oxyben- zoic acid; and red to blue ferric colors are given by brorno and nitro salicylic acids and salicyl-sulphonic acid. (See Car- bolic acid, ferric reaction.) Bromine water gives a crystalline precipitate, C 7 H 4 Br 2 O 3 , very ^slightly soluble in water, freely soluble in alcohol. Solution in 40000 parts of water gives crystals seen under the microscope (ALMEN, 1878). Nitric acid, if concentrated, in the cold, and if dilute, by warming, forms nitro-salicylic acids, then by more intense action forms nitro- phenic (picric) acids, the latter recognized by its intense red- brown color. The reaction is most promptly obtained by Mil- Ion's reagent, fuming acid mercuric nitrate, and gives color in dilution with 1000000 parts of water (ALMEN, 1878). Copper sulphate, with neutral solution of salicylate, gives a green color. Glucose with from two to three times its weight of salicylic acid, the mixture warmed with excess of sulphuric acid (concen- trated), gives a fine blood-red color. Nearly the same color is given by benzoic acid in this test ; a brown to blood-red color by hippuric acid (PmpsoN, 1873). Sodium amalgam, warmed in a slightly acidulated solution, gradually reduces salicylic acid to SALICYLIC ACID. 439 its aldehyde, oil of spiraea, C 6 H 4 .COH.OH, recognized by its odor (compare with Benzoic acid). A mixture of equal volumes of sulphuric acid and methyl alcohol, distilled from a small portion of residue containing salicylic acid or salicylate, yields a distillate odorous of wintergreen oil, methyl salicylate, CH 3 .C 7 H 5 O 3 . Ethyl salicylate, formed in a corresponding way, has a similar odor. Heated with lime, salicylic acid gives the odor of phenol, obtained also by heating salicylates alone (see a). Salicylic acid reduces permanganate of potassium solution, but does not reduce potassium cupric tartrate. Sulphuric acid, not diluted, in contact with salicylic acid at a gentle heat, pro- duces salicyl-sulphonic 4c*. 1884-86: Ber. d. cheat. (?., 17, 2849; 18. 777, 1917; 19, 520; Jour. Chem. Sor.. 48, 276. 819, 1146; 50, 564. *F. L. SoxxEXSCHEor: Ber. d. them. Get., 8, 212; Jour. Chem. Soc., 28, 771. 4 Gmpt. rend., 97, 267; Jour. Chem. Soc., 46, 88. M885: Jour. Chem. Soc., 47, 143. STRYCHNINE. 447 Yield of Strychnos Alkaloids. Total alkaloids : 1.65 to 2.88 per cent. (DRAGENDORFF, 1874 *). More than is generally supposed, the richest specimens reaching nearly 4 per cent. (DL-XSTAX and SHOKT, 1883-85, by their own method*). Dr. A. ti. Lroxs, in 1885,* stated the results of twelve specimens, from 2.68 to 4.89 per cent., giving a mean of 3.16 per cent. Of strychnine alone, 0.96 to 1.39 per cent, in the results of a few lots (DRAGEXDORFF, 1874). As a generally accredited statement, from analyses older than the recent methods, strychnine is found in Ignatius bean as high as 1.5 per cent. ; in nnx-vomica seeds with an average of 0.5 per cent. The statements of Dunstan and Short are given in foot-note below. Methods of analytical separation of strychnine from brucine, in use, are not well assured. Constituents of Nux-vomica, other than the Alkaloids. In combination with the alkaloids, Strychnic or Igasuric Acid, so named, is in fact an iron- greening tannic acid (Honx, Arch, der Phar., 202, 137). A glucoside, Loganin, CggEL^O^, was discovered in nux-vomica by DUNSTAN and SHORT in 1884. 4 In the seeds of Strychnos nux-vomica, and in pharmaceutical prepa- rations made therefrom, it is present in small proportion : in the pulp of the frnit of Strychnos nux-vomica it was found to the extent of 4 or 5 per cent. Loganin, warmed with sulphuric acid, gives a fine red color, which on standing develops into a purple a color-result not unusnal to glucosides. By boiling with dilute sulphuric acid a glucose and a body named logane- tin were formed. STRYCHNINE. C 21 EL^N 2 O 2 = 334. Crystallizes anhydrous. Constitution, p. 446 ; Yield in nux-vomica, given above. Strychnine is identified by the chemical tests of the fading purple, and the crystallization of the dichromate and free al- kaloid, and by the physiological tests of tetanic effect and bit- terness (b and d). Microscopic recognition, a and d. Its limits of quantity are indicated by the limit of response in the fading- purple test (d) and in the physiological tests (b). Solubilities, 1 " Werthbestimmnng," p. 64. * Phar. Jour. Trans. [3] 12, 1055; 15. 157; 4m. Jour. Phar., 55, 467. In the seeds of Cevlon nux-vomica these authors report as follows (Phar. Jour. Trans. [3] 15, 1): No. 1, 1.52 per cent, strychnine, 2.95 per cent, brncine, 4.47 per cent, total " 2, 1.78 " " 3.16 " " 4.94 " 3, 1.71 " " 3.63 " 5.31 " 4, 1.68 " 2.86 " 4.54 " *Proc. Mteh. State Phar. As*oc., 2, 173. *Phn . J.rnr. Trans. [3] 14, 1025; Am. Jour. Phar., 56, 431. 448 STRYCHNOS ALKALOIDS. c, p. 451. Separations (e) by solvents immiscible with water (p. 456), from Nux-vomica (p. 456), from preparations of the latter (p. 457), from Biucine (p. 458), from tissues and foods in cases of poisoning (p. 458), from the urine (p. 460), from alco- holic beverages (p. 400). Limits of recovery from tissues, etc., p. 461. In what organs found in cases of poisoning, J ; how long after death recoverable, e (p. 461). Estimated gravimetri- cally and volumetrically, f. Tests for impurities, g. a. Colorless or transparent octahedra, or needles, or prismat- ic crystals ; or a crystalline white or dull white powder. By spontaneous evaporation of a few drops of an alcoholic solution, on a glass slide, a characteristic microscopic field is obtained, and recognized by comparison with a field from known strych- nine under parallel treatment. The crystals may also be ob- tained on diluting a few drops of the alcoholic solution with particles of water applied to the slide by a pointed glass rod. Strychnine melts at about 300 C. In the " subliming cell" at 221 C. (BLYTH, 1878). A microscopic sublimate of needles is obtained at 169 C. (BLYTH, 1878). Sublimes in part un- changed, giving a sublimate recognized under the microscope (HELWIG, 1864). 1 Strychnine Sulphate, (C 2 JI 22 N 2 O 2 ) 2 H 2 SO 4 . 6H 2 O = 874 (COLEMAN, 1883), is efflorescent in dry air; at about 135 C. melts and (near 200 C., RAMMELSBERG, 1881) parts with its water of crystallization (12.36$). Crystallizes in prisms. Crys- tals with 7H 2 O have been reported, and crystals with 5H 2 O are obtained from alcoholic solution. An acid sulphate, C 21 H 22 N 2 O 2 .H 2 SO 4 .2H 2 O, crystallizes in fine needles. The nitrate, normal, crystallizes anhydrous, in groups of silky nee- dles. The hydrochloride, normal, with 1^H 2 O, crystallizes in soft needles or in prisms, and readily effloresces. ~b. The bitterness of strychnine is stated to be perceptible in a solution diluted to 600000 or 700000 parts. The bitter taste is followed by some degree of metallic after-taste. In effect, strychnine is a tetanic poison, to animals as well as man. Locally it has a very slight degree of irritation. Its tetanic ef- fects are due to its action on the gray nerve-tissue of the spinal cord. It is in some part antagonized by chloral hydrate, aconite, hydrocyanic acid, and nicotine, but these do not serve as anti- dotes to its poisonous action. 1 BLYTH: Jour. Chem. Soc., 33, 316. HELWIG: Zeitsch. anal. Ghem., 3,46. STRYCHNINE. 449 The smallest known fatal dose for an adult person is half a grain. With adults in ordinary varying degrees of susceptibi- lity, the administration of from |- to 2 grains (0.03 to 0.13 gram) is likely to cause death, unless this result be prevented by special conditions or by treatment. Recovery has occurred in cases of poisoning by doses of from 3 to 20 grains, and may occur irre- spective of the quantity taken. The Ph. Germ, places the maxi- mum single medicinal dose of the nitrate at 0.01 gram (-J- grain) ; the maximum daily quantity, 0.02 gram. With frogs Marshall Hall found distinctive effects from the immersion of the animal in a solution of strychnine at a limit of 0.0002 grain (0.000013 gram) ; Harley, by injection into the lungs of very small frogs, obtained spasms from as little as 0.00006 grain (0.000004 gram). By carrying the solution, about 2 grains in amount, into the stomach of a frog (Rana Halecina) fresh from the pond, and of from 15 to 50 grains w eight, Wormley obtained, from 00002 grain (0.000013 gram) of strych- nine, distinctive symptoms in from 10 to 30 minutes ; from 0.002 grain (0.00013 gram), symptoms in 3 or 4 minutes, and death in 15 to 30 minutes ; from 0.02 grain (0.01)13 gram) of strychnine, immediate spasms and death in about 8 minutes. With 0.00007 grain (0.000005 gram) of strychnine, symptoms were obtained in some of the very small animals in 50 minutes ; in other ani- mals no symptoms were obtained. No chemical change of strychnine, in its course through the living body, has as yet been demonstrated. In some part, or at some rate, it may suffer oxidation or conversion in the body, as Plugge and others have believed. In a considerable part, at least, it is in many cases excreted, unchanged, in the urine. In other cases of poisoning, analysis of the urine has failed to reveal it. KRATTER (1882) found strychnine in the urine in half an hour after the administration of \ grain (0.0075 gram) of strych- nine nitrate ; and it continued to be so excreted for 24 hours. When administered several times in succession, it was 3 days after the last ingestion before the alkaloid disappeared from the urine. HAMILTON (New York, 1867) reported the finding of strychnine in the urine on the morning after the patient w T as poisoned. RAUTENFELD (1884, Dor pat) repeatedly obtained strychnine, in crystalline form, from the urine. McADAM found it in the urine of a dog nine minutes after the administration of half a grain, and before symptoms of poisoning appeared. Usually, however, it is not to be found in the urine of animals quickly killed by it. In two cases of dogs, with death after, respectively, 40 and 100 minutes, WORMLEY failed to find the 450 STRYCHNOS ALKALOIDS. poison in the urine. In the liver it is retained to an extent greater than has been found in any other organ (DRAGENDORFF and MASING-, HUSEMANN, ANDERSON). In other organs and in the tissues of poisoned animals its recovery by analysis is very uncertain. When death of the animal very shortly follows the administration, it is in many cases to be found in the blood. DRAGENDORFF concludes from experiments under his direction that strychnine very quickly leaves the blood and becomes re- tained in the liver. 1 G. A. KIRCHMAIER, in experiments made under the observation of the author, found that strychnine was by no means uniformly recovered, even to a qualitative extent, from the blood of animals quickly poisoned with the alkaloid, nor from any of their organs remote from the point of introduc- tion." (As to limits of analysis, in recovery of the alkaloid, see under Separations, e.) '"Meine Erfahrung ilber diesen Gegenstand lasst vermuthen, dass das Strychnin sehr schnell dem BliUe entzogen und in der Leber zuruckgehalten werde, von wo aus es nur sehr langsam wieder in die allgemeine Saftcircula- tion gelangt, urn rait dem Harn aus der Korper entfernt zu werden. Es lasst sich wenigstens bei Hunden und Katzen nicht dafiir einstehen, dass man, selbst wenn der Tod bald nach Darreichung das Giftes erfolgt, im Blute oder den blutreichen Organen (ausschliesslich der Leber) das Gift nachweisen ko'nne. Bei Versuchen, die unter meiner Leitung angestellt wurden, erhielt G. P. Masing bald ein positives, bald ein negatives Resultat, ohne Anhaltspunkte filr eine Erklarung dieser Verschiedenheiten zu gewinnen " (' Ermittelung von Giften," p. 249). 2 Experiments by administration to cats, with dissection and analysis beginning not later than 12 hours after death occurred from the action of 'the poison : G. A. KIRCHMAIER, 1883: Contributions Chem. Lab. Univ. Mich.,2,p. 91. Strychnine given. How administered. Time before deaUi. Dissection. No. 1... ' 2... . 3... . ' 4... . * 5... . ' 6... . One-fourth grain. One-sixth grain. One-eighth grain. One thirty-second grain. One-sixtieth grain. Injection into the back. " " breast. In saphenous vein. By the stomach. 2 minutes. 2* f* 11 20 In % hour. In^ " In 12 hours. At once. In cases Nos. 3 and 4 chloroform was administered before the poison was given. Liver. Kidneys. Blood. Heart. Muscle. Muscle near puncture. Stomach. No. 1. Not found . Not found. Not found. Not found. Not found. Found. Not found. 2. 44 41 44 41 44 44 44 i 4' 44 44 3. Found. Found. Found. 14 < (4 44 4. Not found. Not found. 44 44 5. Found. 6. " " .... ... .... Not found. STRYCHNINE. 451 c. Strychnine is soluble in 67000 parts of water at 15 C. ; in 8333 parts at ordinary temperature (WORMLEY) ; in 2500 parts of boiling water ; in 110 parts of alcohol at 15 O. ; in 207 parts of absolute alcohol or 400 parts of common whiskey (WOKMLEY) ; in 12 parts of boiling alcohol ; in about 500 parts diluted alcohol of sp. gr. 0.941 and in 2617 parts of sp. gr. 0.970 (PRESCOTT and SMITH, 1878) ; in 1400 parts of absolute ether at ordinary tempo rature (WORMLEY), or in 1250 parts of commercial ether (DitA- GENDORFF) ; in 6 to 8 parts of chloroform; in 140 parts of benzene (sp. gr. 0.878) ; slightly soluble in petroleum benzin (DRAGENDORFF), requires 12500 parts (WORMLEY) ; soluble in 122 parts of amyl alcohol; in 300 parts of glycerin (CASS and GARST) ; somewhat soluble in certain essential oils ; sparingly soluble in ammonia-water, not soluble in solutions of fixed alka- lies. Fine octahedral crystals are obtained from the benzene so- lution. Strychnine in alcoholic solution gives a decided alkaline reac- tion to test-papers ; and it forms stable salts, mostly of a neutral reaction. Strychnine sulphate (a, p. 448) is soluble in 42.7 parts of water at 15 C. (COLEMAN, 1883) ; very freely soluble in boil- ing water ; in 60 parts of alcohol at ordinary temperature, or 2 parts boiling alcohol ; insoluble in ether, or chloroform, or ben- zene, or amyl alcohol ; soluble in 26 parts of glycerin. /Strych- nine nitrate is soluble (Ph. Germ.) in 90 parts of cold or 3 parts of boiling water, and in 70 parts of cold or 5 parts of boiling alco- hol. The hydrochloride is soluble in 50 parts of cold water. d. Qualitative tests. The fading purple : If strychnine or one of its ordinary salts, purified from non-alkaloidal matter, in a film of dry residue or a particle of dry mass, on a white porce- lain surface, be moistened with pure concentrated sulphuric acid, in the cold, no coloration occurs. If now a just visible fragment of crystallized potassium dichromate, taken on the end of a nar- row glass rod, be placed for a moment in the moistened film of the test material, and then drawn out through it, a distinct purple to blue color appears, soon changing to reddish yellow tints, and fading away into the slight colors due to the dichromate itself. The area of moistened film, taken at first, need not be over a fourth of an inch in diameter, and the liquid is drawn in one direction only, toward the side of the dish, as the dichromate is carried through it. in repetition of the trial. The reaction is obtained by the sulphuric acid and an oxidizing agent, and lead peroxide, ceroso-ceric oxide, manganic hydroxide, potassium per- manganate, and potassium ferricyanide have severally been used 452 STRYCHNOS ALKALOIDS. for this purpose. 1 The permanganate, cerium oxide, and ferri- cyanide are usually mixed with the sulphuric acid before the test : one part of permanganate in 2000 parts of the acid, etc. But if this be done the film must be separately tested with sulphuric acid alone. The proportion of oxidizing agent must be minute in testing for minute quantities of the alkaloid. The color given by the oxidizing agent itself, in the sulphuric acid, must be observed. If traces of non-alkaloidal matters are present, a portion of similar matters is subjected to the same test, and any shades of color developed are to be taken into the account. Dichromate in sulphuric acid has a slight color, from the yellowish-red of the dichromate itself to the chromic sulphate formed by reduction. Permanganate presently becomes green in sulphuric acid. These tints do not resemble the fading pur- ple, and in use of the proper minute quantities of the oxidizing agent they do not obscure the strychnine reaction, or not until the extreme limit of recognition of the latter is reached. It is to be remembered that the evidence of strychnine depends upon the joining of three results : (1) no coloration by the sulphuric acid, (2) a blue or purple color when the oxidizing agent takes effect, (3) the fading of the blue or purple color. The use of a good hand-magnifier adds efficiency to the test, but all the re- sults should be unmistakable to the eye without special aids. By the manipulation with the dichromate as above given in detail, and applied to a residue from 1 c.c. of solution, a good and strong color can be obtained from 0.0000025 gram (0.000037 grain) of strychnine. 3 1 As to special effects of vanadic acid in this test, MANDELIN, 1883. As to a certain product of this oxidation, see p. 446. 2 In detailed experiments the following results were obtained (G. A. KIRCHMAIER, 1883: Contributions Chem. Laboratory Univ. of Mich., 2, 89; A. B. PRESCOTT, 1885: "Control Analyses and Limits of Recovery," Chem. News, 53, 78): Experiment. C.c. strychnine solution evaporated. Strychnine sulphate in grams. The fading purple. No 1 5.0 0.0000125 Distinct. 2 40 0.00001 3 30 0.0000075 < 4 2.0 0.000005 K 5 1.0 0.0000025 Good. Play of colors 6 0.8 0.000002 less marked. Faint. 7 0.6 0.0000015 Uncertain. The solution was evaporated in a common evaporating-dish. Doubtless STRYCHNINE. 453 WORMLEY ' obtained very satisfactory evidence, by the use of the dichromate, from 0.0000013 gram (0.00002 grain) of strych- nine ; 0.000007 gram giving evidence as satisfactory as could be obtained from any larger quantity, and 0.0000007 gram, when deposited within a narrow compass, giving a distinct coloration. S. J. HINSDALE (1885) prefers the ceroso eerie oxide as an oxi- dizing agent, and reports a good play of colors from the 0.0000007 gram (0.00001 grain) of the alkaloid. It is to be understood that the limit of delicacy of the color test is wholly dependent upon the concentration of the alkaloid. A barely visible fragment of crystal of strychnine gives a good play of colors, but if dissolved in a few c.c. of alcohol, and the solution evaporated in a common dish, the residue would give probably a negative, possibly an uncertain, result. These statements of the smallest quantity of pure and unmixed alkaloid capable of iden- tification give no answer whatever to the question as to the smallest quantity of the poison, existing in a stomach or a por- tion of food or a plant, capable of recovery and identification. The limits of recovery receive attention, with methods of Sepa- ration, under 0, p. 461. Interferences with the color test (1) Substances diminishing the delicacy of the reaction. Brucine in equal quantity with strychnine prevents the coloration, unless the quantity of each be very minute, less than 0.001 grain, but a mixture of 0001 grain of each gives satisfactory evidence of strychnine (WORM- LEY). " The 0.01 grain of strychnine with 0.001 grain of brucine yields a very marked reaction, although somewhat masked " (Ibid.) Morphine is nearly as influential as brucine in dimin- ishing or preventing the color test. A residue from a solution of 01 grain each of strychnine and morphine gave WORMLEY 2 little indication of the presence of strychnine ; but a similar mix- ture of 0.001 grain of each of these alkaloids gave good evidence of strychnine ; while even a minute quantity of a mixture of three parts morphine to one part of strychnine gave a negative result. The absence of both these alkaloids, therefore, should be assured, if need be, by use of separative solvents, as directed un- der Separations (e). Of inorganic salts, nitrates and chlorides have been named as diminishing the reaction. Organic matters had the residue from the 0.8 c.c. of No. 6, or even that from the 0.6 c.c. of No. 7, been brought within an area of two or three millimeters diameter, and moist- ened with much less than a drop of sulphuric acid, a good play of colors would have been obtained in trials 6 and 7. 1 " Micro-chemistry of Poisons," 1885, p. 564. 2 1860: Chem. News, i, 243. 454 STRYCHNOS ALKALOIDS. acting as reducing agents undoubtedly hinder or prevent the reaction, and sugar is especially influential in this regard. While it is the rule that the test is to be applied in the absence of sub- stances not alkaloids, in practice it is sometimes difficult to be certain whether these matters are present or not. This question can be decided by a control-test as follows : Obtain by itself a narrow film of residue, equal to that tested for the result of analysis, by evaporation of a little of the recovered solution in a porcelain dish by itself. Add thereto (aside from the portions under analysis) say 1 c.c. of a solution containing in each c.c. from 0.0000025 to 0.000005 gram of strychnine sulphate (p. 452), and evaporate again. Or evaporate the 1 c.c. with the small portion of solution under analysis. This residue, with the added known quantity of strychnine, should give a distinct fading- blue coloration, as distinct as can be obtained by a test upon a residue from 1 c.c. of the strychnine solution unmixed. (2) Substances giving, in part, the same results obtained from strychnine in the fading-purple test, and presenting the so-called "fallacies" of this test. The greater number of these substances give a color with sulphuric acid alone, and therefore their results are at once excluded from all indication of strychnine. Among these sub- stances may be named papaverine, thebaine, cryptopine, ber- berine, amygdalin, veratrine, and cod-liver oil. Aloin gives a greenish color, fading to yellow. Aniline, colorless with sulphu- ric acid alone, on adding the oxidizing agent presents yellowish or greenish tints slowly deepening to blue, which deepens, and (instead of fading) finally becomes blue-black to black. Gelse- mine, colorless alone with sulphuric acid, on adding dichromate or other oxidizing agent gives a reddish-purple to cherry-red color, somewhat resembling that of strychnine. Hydrastine, but faintly yellowish with sulphuric acid, on adding the dichromate gives red to green color (LYONS, 1886). Curarine, the non-crys- tallizable principle of worara, obtained from botanical sources allied to those of strychnine, is the only substance besides strych- nine which has been found to give the threefold result of the fading-purple test (WORMLEY). The use of the permanganate, as an oxidizing agent, is more exposed to fallacy than the other oxidizing agents. If the oxidizing agent be mixed witli the sul- phuric acid, and no parallel trial be made with sulphuric acid alone, the reaction of cod liver oil may well be assumed as an indication of strychnine. The physiological test is to be placed second in order of the value of evidence. The data for this test with the frog, and the limits of quantity revealed by it, are given under b, p. 449. For STRYCHNINE. 455 the test the alkaloid is obtained in a neutral aqueous solution of a salt, as the sulphate. The taste of a graded dilute solution gives corroborative proof as to the presence and limit of quantity of strychnine (5, p. 448). According to WORMLEY, a grain of a l-50000th solution of the alkaloid unmixed with other matters has a quite perceptible bitter taste ; and a drop of a l-10000th solution, even in mixture with a very notable quantity of foreign matter, usually has a decided bitter taste. Potassium dichromate solution, added to solutions of strych- nine salts not very dilute, gives a crystalline yellow precipi- tate of strychnine dichromate, (C 21 H 22 N 2 O 2 ) 2 H 2 Cr 2 O 7 (DrrzLEK, 1886), its slow formation being promoted by stirring. The crys- tals include octahedra and often bush-like groups. A drop of the solution, on a glass slide, may be treated with a drop of the dilute reagent, the mixture stirred with a fine pointed glass rod, -and from time to time examined under a microscope with a low power. The precipitate is not soluble in excess of the reagent or in quite dilute acids. Solutions of strychnine salts in 1000 parts water do not yield an immediate precipitate, but from this .and much more dilute solutions crystals can be obtained as di- rected above. The general reagents for alkaloids give precipitates of strych- nine. The precipitate with Mayer's solution, potassium mer- curic iodide, appears in a solution of a salt of the alkaloid in 150000 parts of water. The precipitate by phosphomolyb- date dissolves in ammonia without coloration. The precipitate by iodine in potassium iodide solution is obtained in very di- lute aqueous solutions (1 : 100000), reddish-brown, and soluble in alcohol. From the alcoholic solution somewhat characteristic crystals can be obtained. Alkali hydrates give crystallizable precipitates, soluble in excess only in the instance of ammonia. The ammoniacal solution gives fine crystals of the free alkaloid <, p. 448). < Strychnine is noted, among alkaloids, for its stability under ordinary influences of decomposition. By action of chlorine or bromine, monochlorstrychnine or bromstrychnine is readily formed, as a substitution compound : by action of iodine, iodated hydriodides are formed, as addition compounds. By treatment with methyl iodide, the salt of methyl-strychnine, C 21 H 21 (CH 3 ) N 2 O 2 .HI, is easily obtained, as also is ethyl-strychnine by the same means. The conversion of strychnine into brucine remains under investigation. See Constitution of strychnos alkaloids, p. 446. 456 STRYCHNOS ALKALOIDS. e. Separations. Strychnine may be concentrated by evapo- ration of its solutions at 100 C., without loss or decomposition. In separation by solvents immiscible with water, chloroform and benzene take it up most abundantly as a free alkaloid (from al- kaline aqueous solutions). According to Dragendorff, petroleum benzin, though dissolving strychnine but very sparingly, may be profitably used to take it up from alkaline solutions as a means of [qualitative] separation from alkaloids soluble in chloroform or benzene and not soluble in petroleum benzin. 1 From acidu- lous solutions strychnine is not taken by any of the ordinary sol- vents immiscible with water (except as traces of the aqueous solution itself may be carried in solution with ether, chloroform, and amyl alcohol). From the Nux-vomica, in total alkaloids. The method of Messrs. DUNSTAN and SHORT," which has met with general appro- val, is as follows : Of the finely powdered nux-vomica seeds 5 grams are packed in the percolator of a continuous extraction apparatus, and treated actively with 40 c.c. of alcoholic chloro- form containing 25 per cent, of alcohol, until exhausted, which is usually accomplished in two hours or less. The chloroformic solution is agitated (in a separator) with 25 c.c. of a ten per cent, diluted sulphuric acid, the layer of chloroform drawn off and shaken again with 15 c.c. of the diluted acid, and the chloro- form layer drawn off. The formation of the chloroform layer ie- much facilitated by gently warming the mixture. The mixed acid solutions should be quite free from undissolved chloroform, and entirely clear. Chloroformic turbidity may be removed by adding a little chloroform and agitating slowly by gradually in- verting the separator. If need be, the mixed acid solutions- should be filtered, through a filter wet with the dilute acid, and the filter washed with a very little of the dilute acid. The total acidulous watery solution is now made alkaline with ammonia, and shaken out, in the separator, with 25 c.c. of chloroform. The clear chloroformic layer is slowly drawn off into a weighed or balanced beaker. If not readily obtained clear by subsiding, the chloroformic solution may be run through a small double filter wet with chloroform, washing the filter with a little chloro- form. The chloroform is gently evaporated, a constant weight 1 Wormley states that strychnine requires 12500 parts petroleum benzin for solution. Even if this hold good for the alkaloid freshly liberated, it still would require only 0.1 c.c. of the solvent to carry a quantity of the alkaloid easily identified by the color test. 2 1883: Phar. Jour. Trans. [3] 13, 665. Given here with very slight addi- tions in details. STRYCHNINE. 457 of residue is obtained at 100 C. or on the water-bath, for which one hour is usually enough, and the weight of residue taken, for the quantity of total alkaloids. From preparations of Nux-vomica, in total alkaloids. DUNSTAN and SHOUT present directions substantially as follows for standardizing an alcoholic percolate of nux-vomica in prepa- ration of a fluid extract of uniform alkaloidal strength. 1 The operation may be applied to the medicinal tincture or fluid extract. Take of the liquid 25 c.c., or one fluid-ounce, or other suitable quantity by weight or volume, according to the purpose. Eva- porate nearly to dryness over the water-bath. Treat the residue with water acidulated with sulphuric acid, in the proportion of 30 c.c. of a 7.5 per cent, sulphuric acid for each 6 to 7 grams of nux-vomica represented (1 f. oz. for each 100 grains), adding at the same time, for same quantities, about 7 c.c. (2 fluid - drachms) of chloroform. Agitate and warm gently. When the chloroformic layer has separated draw it off, add to the aqueous liquid ammonia to cause an alkaline reaction, and agitate with 15 c.c. (or \ f. oz.) of chloroform, warming as before. Draw off the chloroformic solution into a weighed dish, evaporate, dry over a water-bath for one hour, cool, and weigh for total alka- loids. For the Solid Extract the same authors dissolve 10 grains (or 0.6 gram) in J f. oz. (15 c.c.) of water, with the aid of heat,, add 60 grains (4 grams) of sodium carbonate previously dissolv- ed in f. oz. (15 c.c.) of water, and agitate with f. oz. (15 c.c.) of chloroform, warming to obtain a separation. The chlorofor- mic solution of alkaloids is carefully drawn off, and agitated with f. oz. each of diluted sulphuric acid and water (or 30 c.c. of 5 to 7 per cent, sulphuric acid). The clear acidulous watery solution is made alkaline by adding ammonia, and agi- tated with f. oz. (15 c.c.) of chloroform. When the liquids have separated the chloroform is evaporated off in a weighed dish, the residue dried for an hour over the water-bath and weighed as total alkaloids. In applying this process to the resi- due from (say 25 c.c. of) the alcoholic preparations, Dr. A. B. LYONS 2 directs to shake out the acidulous liquid, first, with two successive portions of ether (25 c.c.), then with one portion of a mixture of one volume of chloroform with three volumes of ether, the shaking not to be too violent. The aqueous* liquid made alkaline is extracted by the same ether-chloro- 1 1884: Phar. Jour. Trans. [3] 13. '1885: Proc. Mich. State P/iar. Asso., 2, 183. 458 STRYCHNOS ALKALOIDS. form mixture, applying it in two successive portions (30 and 20 c.c.) Separation of Strychnine from Brucine. - (1) By dilute al- cohol of sp. gr. 0.970 (about 21$ weight, 26$ vol.) In 1878 ' the author communicated results as follows : Solution requires For strychnine, 2617 parts of 21$ (weight) alcohol, 500 parts of 39$ alcohol. For brucine, 38 parts of 21$ (weight) alcohol, 22 parts of 39$ alcohol. When one part each of strychnine and brucine were digested one hour at ordinary temperature with 100 parts of alcohol of sp. gr. 0.970, filtered, and the undissolved alkaloid washed with 100 parts of the same alcohol, the residue of strychnine gave no qualitative test for brucine, but the brucine left on evaporating the filtrate gave a slight color test for strychnine (even in pre- sence of the excess of the brucine). (2) By precipitation with ferrocyanifJe (DUNSTAN and SHORT, 1883). The sulphates of the alkaloids in aqueous solu- tion acidified with sulphuric acid are precipitated with potas- sium ferrocyanide. The strychnine is entirely precipitated, both when alone and when in the presence -of brucine, while the hm- cine is not precipitated unless in concentrated solution. ScHWEit- tsiNGEE (1885) did not obtain good results by this method. Separation from.' Tissues and Foods in analysis for poi- sons. A weighed quantity (one part) of the material is placed in an evaporating-dish or wide beaker, finely divided under the points of a pair of sharp, bright shears, with moistening if need be to bring to a soft and homogeneous pulp, about an equal quantity of w r ater containing about one per cent, of sulphuric acid is added, and the mass digested, with stirring, on the water- bath for 15 minutes. Four or five parts of well-rectified 90$ al- cohol are added, and the whole digested, with frequent stirring, at a little below the boiling temperature of the alcohol, for about an hour. The edges are to be kept free from dried residue. It is then cooled and strained through close muslin, or filtered through open filter-paper by the help of a filter-pump. The residue is digested, successively, with two smaller portions of alcohol, keeping the reaction of the mass constantly acid, the filtrates from all being received together in a wide mouthed flask, and the strainers or filters well washed with the alcohol. The filtered liquids are concentrated to a syrupy consistence, with 1 Pro. Am. Pharm., 26, 806; Year-book of Phar., 1879, 97. STRYCHNINE. 459 occasional gentle rotation of the flask, preventing dried residues on the edges. Four or live parts of absolute or nearly abso- lute alcohol are added, the mixture shaken by rotating the flask, and, when cold, filtered into another flask, arid the filter well washed with the absolute alcohol. The entire filtrate is evaporated on the water-bath to remove all the alcohol, when about two parts of water are added. If the reaction be not sharply acid to litmus, it is made so by adding a drop or two of diluted sulphuric acid. The mixture is filtered, and the resi- due and filter well washed with small portions of water, re- ceiving the entire filtrate in a small separator, or strong tube having a good cork. It is now shaken out with chloroform, in one or more portions, or as long as this solvent continues to ex- tract anything. The clear chloroform layer is drawn off by the separator, or forced out of the tube as water is delivered from a wash-bottle, the tubulated stopper fitted for that purpose having an adjustable delivery tube brought to the conical bottom of the container. The chloroform solution is washed once or twice with a little faintly acidulous water, and the washings added to the aqueous liquid. The chloroform solution is reserved for any tests desired. The aqueous solution is made distinctly alkalin^ by adding ammonia, and exhausted by shaking out with from three to five portions of the chloroform. The united chloro- formic liquids are to be obtained perfectly clear. A zone of par- tial emulsion next to a supernatent watery layer may be resolved by introducing into the zone a c.c. or so each of fresh chloro- form and of water, and tapping the separator. If need be, the chloroform so added may be made slightly alcoholic. Also, a little portion of emulsified chloroform may be washed with chlo- roform on a filter wet with the same solvent. Now an aliquot part of the total chloroformic solution (from the alkaline liquid) may be evaporated for preliminary tests. The residue so ob- tained is usually colored and loaded with substances not alka- loids. When no other alkaloid than strychnine is sought, the residue from evaporation of the (remaining) chloroform solution may be purified as follows : The residue is treated with concen- trated sulphuric acid, one or two drops, or only enough to moisten the whole, covered and heated on the water-bath, or, better, heated in an oven at 100 C., avoiding any higher tempe- rature, for an hour or so. To the carbonized mass, when cold, barium carbonate is added, to neutralize nearly all the acid, still leaving an acid reaction. The little mass is now exhausted with small portions of water, and the solution and washings filtered into a small separator. The aqueous solution is now made alka- 460 STRYCHNOS ALKALOIDS. line by adding ammonia, and exhausted by repeatedly shaking out with chloroform in small portions. The entire chloroform solution is received in a graduated cylinder, and aliquot parts are evaporated in small porcelain and glass dishes, for the several tests, and for trial as to qualitative limits, also, if desired, for volumetric estimation. The residues will contain some ammo- nium sulphate, the crystals of which are likely to be seen in mi- croscopic examinations. To obtain a portion free from ammo- nium salts and from sulphates in general, treat one of the residues with water and barium carbonate, evaporate to dryness, take up with warm alcohol, filter, and evaporate the alcoholic solution, for another residue. Strychnine may be separated from the Urine by evaporat- ing the acidulated liquid to a syrup, stirring with strong or abso- lute alcohol, filtering and washing well with the same alcohol, concentrating the filtrate to a syrup, which is dissolved in water enough to make a limpid solution. This is washed (while aci- dulous) with chloroform ; then made alkaline by adding ammo- nia, and washed with one or more portions of chloroform. The chloroformic solution is evaporated, either all together or in aliquot portions, for direct tests upon the residue or for further purification, as directed on p. 459. As to the occurrence of strychnine in the urine, following administration, results are stated under 5, p. 449. Recovery from Alcoholic Beverages. Messrs. GRAHAM and HOFMANN, in 1852, made extensive analyses of the ales and beers of Great Britain for strychnine, as follows : Half a gallon of the ale was shaken with 2 oz. of animal charcoal, and, after standing 12 to 24 hours, filtered through paper. The drained charcoal was boiled half an hour in purified alcohol, and the fil- tered alcohol distilled off. The watery residue was made alka- line with potassa, and agitated with an ounce of ether [chloro form]. The residue from evaporation of the ether was tested. Taking -| grain of the alkaloid in J gallon of the malt liquor, the operators invariably obtained the color test in the final residue. Probably a preferable procedure would be the treatment directed above for separation from the urine. With distilled spirits, such as whisky, the same operation (last referred to) is advisory, the alcohol being first distilled off, when the slight quantity of resi- due usually renders the operation a very simple one. Single cases of the detection of strychnine in beer in Eastern Europe have been reported. There is no evidence, and no probability, that strychnine has ever been used in the sophistications of dis- tilled spirits. STRYCHNINE. 461 Limits of Recovery from complex orgdnic matter. From the experiments of Mr. Kirchmaier, with the author's co-opera- tion, 1 it appeared that the limits of analytical separation, by a process essentially the same as that just detailed, were as follows : With 50 grams of meat, the loss of strychnine was for one part of meat, 0.00000795 part of strychnine ; or for one part of the alkaloid, 125786 parts of the tissue-material. 2 That is to say, in separating strychnine from an avoirdupois pound of tissues the loss of alkaloid is from one thousand to two thousand times the least quantity needed for certain identification (p. 452). The deposition of strychnine in various organs of the living body receives statement under &, at p. 450. It is capable of recovery from partly decomposed organs some time after death, subject to limitations not yet very definitely established. No other alkaloidal poison resists destruction in the interred body any better than strychnine, probably none other resists as well. Yet it is by no means indestructible when contained in putrefactive tissue. The data obtained by adding strychnine to tissues, and recovering it after some months of putrefactive decomposition, are but imperfectly applicable to cases of poisoning by the al- kaloid, because of the attenuation that results from absorption and elimination. Wormley states 3 that " the longest period in which the analysis furnished positive evidence of its presence in the exhumed human body is 43 days after death (Ann. d?Hyg., 1881, 359)." But in the case of Magoon, in New Hampshire, 1875, Drs. Hayes and E. S. Wood, of Boston, found strychnine in the body of a woman advanced in years, exhumed one year and three days after death; and the analysts reported 0.15 grain in the stomach, 0.23 grain in the liver, and presence of the alka- loid in the intestines. Death had occurred in less than an hour after administration of an unknown quantity of the poi- 1 "Control Analyses and Limits of Recovery in Chemical Separations,' 1 1885: Chem. News, 53, 78. Contributions from the Ghem. Lab., Univ. of Mich., 2,91. 2 Experiment 4, 5 grams meat, 0.00016 strychnine, good color-test. 5, 0.00012 faint 6, 0.00008 very faint 7, 0.000064 negative * 8, 5 0.0004 good 9, 0.0002 faint ' 10, 0.00016 negative 3, br ad, 0.0001 good 4, 0.0008 faint 5, 0.0006 very faint r * rt 1 _ T ' 6, __ ^ 0.0004 negative '* \ 3 2ded. " Micro-Chem. Poisons," 1885. 462 STRYCHNOS ALKALOIDS. son, in a tumbler of hot liquid of extreme bitterness. In 1875 Prof. WORMLEY made analysis of the stomach and liver seven months after death, and the chemical tests gave no evidence of strychnine, although the final residues had a bitter taste. Death had occurred within two hours, and the symptoms and other proofs were such that there was a conviction for poisoning (Ohio v. Dresbach, 1881). Buchner, Gorup-Besanez, Wislicenus, and Eanke (1881) made a series of experiments upon seventeen dogs, killed, each, by 0.1 gram of strychnine, and buried from 100 to 330 days before analysis. In no case did the chemical tests show the presence of strychnine, though the physiological test, with frogs, obtained tetanic convulsions, and the residues had a bitter taste. f. Quantitative. Strychnine is estimated gravimetrically by weight of the anhydrous alkaloid, dried at 100 C. A volu- metric estimation (less exact than the gravimetric) is obtained by Mayer's solution, each c.c. of which represents 0.0167 gram of strychnine (^^^^ of 334, in grams). The end of the reaction is distinct, and the precipitate settles fairly well in acidulated water, but settles better in a concentrated solution of potassium chloride (DRAGENDORFF, 1874), when each c.c. of the entire solu- tion dissolves 0.00216 gram of the precipitate (ibid.) The composition of the precipitate as C^H^KgO^HIHglo was near- ly sustained by the determinations of mercury and of iodine communicated by the author in 1880. 1 This chemical formula corresponds to 36.47$ strychnine in the precipitate. Dragendorff gives a gravimetric trial by washing, drying, and weighing the precipitate, whereby there was a loss of only 1.8$ on the basis of this formula. Though more constant than the greater number of alkaloidal iodomercurates, the precipitate is not the most favora- ble form for gravimetric purposes. And in the volumetric de- termination the solution is to be made of 200 parts to 1 of the alkaloid. g. Tests for Impurities. " Strychnine should not be red- dened at all, or at most but very faintly, by nitric acid (absence of more than traces of brucine)." May be colored yellowish but not red when rubbed with nitric acid (Ph. Germ.) Not colored by nitric acid (Br. Ph.) A more strict exclusion of brucine is effected by washing the free alkaloid with dilute alcohol (sp. gr. 0.970) to separate the brucine, as described on p. 458, the residue 1 CHEM. LAB. UNIV. MICH.: Am. Chem. Jour., 2, 297-99; Jour. Chem. Soc. 42, 664. BRUCINE. 463 from evaporation of the filtered dilute alcohol being tested with nitric acid. 1 Of ten samples of commercial strychnine, treated in this way, all but two gave tests for brucine. BRUCINE, C 23 H 26 ]SroO 4 = 394. Crystallizes with 4H 2 O, 15.45$. For constitution of the alkaloid see p. 446 ; yield in nux-vomica, p. 447. Brucine is identified by its color-tests with nitric acid and other additions, its dichromate precipitate, its crystalline forms, and, its physiological effects as a convulsent (d). It is distin- guished from strychnine by a negative result in the " fading- purple " test, and the positive reactions with nitric acid, etc. (d). It is separated with strychnine, and from strychnine, as de- scribed on pages 456, 460 ; and is estimated gravimetrically or volumetrically (/") a. Transparent or colorless, oblique, four-sided crystals, or in groups of delicate needles, varied in form according to the solvent and the conditions. The crystals effloresce in dry air, and on the water-bath the alkaloid soon becomes anhydrous. It melts at 151 C. (BLYTH, 1878), at 115.5 C. [when anhydrous?] (Gur). It gives a decomposition-sublimate in the " subliming cell" at 150 C. and above (BLYTH), at 204 C. (Guy). ~b. Brucine is extremely bitter. In effects in general it re- sembles strychnine, but a far greater quantity is required for the same effect T. L. BKUNTON (1885) found that it is excreted far more rapidly than strychnine, so rapidly that when given by the stomach to animals pure brucine has little effect. Given hypo- dermically it causes death by convulsent action. WOKMLEY states that the effect of brucine is that of strychnine, with T ^ the intensity. c. Effloresced brucine dissolves in 850 parts cold or 500 parts boiling water, the crystals being considerably more soluble. Very soluble in alcohol, absolute or aqueous. As to its solubility in certain strengths of dilute alcohol, see page 458. Almost insoluble in ether, soluble in chloroform, benzene, or amyl alcohol. The ordinary salts of brucine are soluble in water and in alcohol, not in ether. d. Nitric acid gives a red color with brucine. For the proper intensity the acid should be concentrated, near 1.42 spe- cific gravity, and the alkaloid should be dry and placed over a 1 The author and A. D. Smith, 1878: Proc. Am. Pharm., 26, 807. 464 STRYCHNOS ALKALOIDS. white ground. If the alkaloid be concentrated at one point, and minute in quantity, it may be treated with less than a drop of the acid, added at the point of a sharp glass rod. On standing, or heating, the color changes to yellowish ; on evaporating to dryness the red color returns in the residue. About 0.0000013 gram (0.0000-2 grain) is the limit of quantity for distinct colora- tion, with the best concentration. Sulphuric acid alone applied to the dry alkaloid causes a faint rose color. If in a drop of the rose solution of the concentrated acid a minute fragment of po- tassium nitrate be placed, an orange-red color is obtained. If the concentration be of the best, about the 0.00003 gram is suf- ficient for a sensible reaction. If the dry alkaloid or its salt be treated with a drop or just wet with nitric acid, as above direct- ed, warmed till the color turns to the yellow, then cooled and touched with a drop or less of good solution of stannous chlo- ride, a purple to violet color is obtained. Excess of either the nitric acid or the tin salt is to be avoided. The heating is only necessary to bring out the full delicacy of the reaction. Sodium sulphide solution (by saturating caustic soda solution with H 2 S) may be used instead of stannous chloride. The reaction with tin salt may be recognized with the 0.00001 gram of the alkaloid. Of the three allied color-tests just described, the last is the most characteristic, and the agreement of the three furnishes quite conclusive proof of identity, with distinctions from morphine, narcotine, and other alkaloids. In the sulphuric acid and di- chromate test made for strychnine, brucine slowly reduces the chromic acid, with colors changing from dull orange to greenish, without the least resemblance to the " fading purple." Froehde's reagent gives a red to yellow color. A solution of brucine in dilute sulphuric acid, touched with very dilute dichromate solution, gives a red color changing to duller tints. Mer- curous nitrate (free from excess of nitric acid) gives a re- action somewhat like that of nitric acid, but developed only on heating, the color being carmine, and permanent on evaporating to dryness. Solution of a brucine salt, with solution of potassium dichro- mate, yields a yellow crystalline precipitate of brucine chromate, in groups of bent needles, formed in quite dilute solutions, and somewhat characteristic. The precipitate dissolves in nitric acid with a red color. The general reagents for alkaloids give the customary preci- pitates with brucine. Very dilute solutions give the precipitate with iodine in potassium iodide solution. The precipitate form- ed by phosphomolybdate is of an orange tint, dissolving in am- TANNINS. 465 inonia to a yellow-green solution. The caustic alkalies cause a precipitate, gradually becoming crystalline, and somewhat sol- uble in ammonia. The physiological test of brucine, with the frog, is qualita- tively nearly the same as that for strychnine (pp. 454, 449), but a very much larger quantity of brucine is required for the same effect. e. Separations. Brucine may be obtained from an aqueous or other solution by evaporation at 100 C., without loss. The aqueous solution of its salts may be washed with any of the ordinary solvents immiscible with water, its salts not being solu- ble in these solvents. On making the aqueous liquid alkaline, chloroform or benzene serves well as a solvent, and amyl alco- hol also takes it up. Petroleum benzin dissolves it to some extent. The separation of brucine with strychnine, from nux-vomica and from its preparations, is described under Strychnine, p. 456. The separation from strychnine is given at p. 458. In analysis for poisons, brucine will be obtained with strychnine by the methods detailed at pages 458, 460. f. Quantitative. Brucine is estimated gravimetrically in the same manner as strychnine (p. 462), the residue of free al- kaloid being dried at 100 C. to a constant weight, when it can be weighed as anhydrous In the volumetric method with May- er's solution, Mayer's factor for 1 c.c. of the solution was 0.0233 gram of the alkaloid. SULPHOCARBOLIC ACID. See PHENOL, p. 405. TANNINS. Tannic Acids. Gerbsauren. -- Vegetable educts of an astringent taste, amorphous or obscurely crystal- line solids, not volatile without change, of very slight acid power, and freely soluble in water and in alcohol. They give blue or green precipitates with ferric salts, and thick precipitates with gelatin, albumen, and starch paste. In most cases they pre- cipitate the alkaloids, likewise tartrate of antimony and potas- sium, and are dissolved but sparingly by dilute mineral acids. They are all strong reducing agents, giving reductions with Fehling's solution, with permanganate, and with salts of silver and of gold. The greater number of them convert animal mem- brane into leather. They are darkened and decomposed by al- kali hydrates. In solutions they are instable. There are many diversities of character and composition of 4 66 TANNINS. The best known of these differences may be stated as tannins, follows : (1) Gluooside-tannins. When boiled with dilute mineral acids, yield (a) a crystalliza- ble acid or its anhydride, or (b) a phlobaphene (a re- sin-like body), along with a glucose. 1 (2) Iron-bluing tannins. With ferric salts give blue to black precipitates or colors. The ferroso-ferric solutions, slightly basic, give the best reactions. Mineral acids dissolve and decolor. (3) Tannins not tanning agents. Do not form leather, nor preserve animal membrane, though precipitating solu- tions of gelatin (WAGNER 3 ). (4) Tannins which, in sublim- ing, or in fusing with po- tassium hydrate, yield a trihydroxybenzene, * C 6 H 3 (1) Tannins not glucosides. For determination whether a glucoside or not, see be- low. (2) Iron - greening tannins.* With ferric (basic) salts give greenish precipitates or colors. Brown colors sometimes obtained. Tints varied by conditions. (3) Tanning materials. Change animal membrane into leath- er, not putrescible. Also precipitate solutions of gela- tin. (4) Tannins which, in sublim- ing, yield a dihydroxy phe- nol, C 6 H 4 (OH)2, and, on fusing with potash, yield an 1 " I have arrived at the curious result that tannic acid, when acted upon by acids, yields, together with gallic acid, sugar, so that henceforth tannic acid maybe classed with the conjugate sugar compounds." STRECKER in a letter to Hofmann in 1853. In 1872 SCHIFF found the product to be primarily digallic instead of gallic acid. 2 Of the iron-greening tannins examined only willow-tannin was found to be a glucoside. STENHOUSE, 1861. " Tannins in the green parts of plants, ac- cording to their nature, affect iron solutions differently ; that which colors iron green is apparently an oxidation product of that which colors iron blue, and the author thinks that the latter, under the influence of transpiration, breaks up into the former modification and sugar/' E. JOHANSON, 1879: Jour. Chem. Soc., 26, 161, from Arch. Pharm. [3] 13, 103-130. Regarding iron colors with phenol hydroxyl, see SCHIFF as quoted under Carbolic Acid, in reaction with Ferric Chloride, p. 399. 3 Zeitsch. analyt. Chem., 5, 1. Wagner states that the pathological tan- nins of the galls of species of Quercus and Rhus do not form true leather or preserve animal membrane from putrefaction. LOWENTHAL (1877: Zeitsch. anal. Chem,^l6, 47) found that precipitates of gelatin with gallotannin and sumach-tannin, standing under water for two years, still gave no odor of putre- faction. TANNINS. 467 (OH) 3 , such as Pyrogallol acid, as Protocatechuic acid. (HLASIWETZ). C 6 H 3 (OH) 2 CO 2 H (HLASI- WETZ). (5) Pathological tannins. (5) Physiological tannins. Formed in punctured vege- From uninjured vegetable table tissues. Gallotanmns tissues (Wagner). Include (WAGNER, 1866 '). Includ- various glucosides and iron- ing sumach-tannin (STEN- bluing tannins. HOUSE, 1861). In testing for glucoside tannins, the solution, not very di- lute, is iirst tried as to its deportment with ferroso-ferric solution, gelatin solution, cinchona sulphate solution, and Fehling's solu- tion. Sulphuric acid equal to one or two per cent, is now added to a portion, and the liquid is boiled for an hour or two Or hydrochloric acid is added, to give about the same percentage of real acid, the liquid sealed up in glass tubes and heated at 100 C. for an hour or longer. A portion of the liquid is now dropped into cold water, to see whether sparingly soluble fer- mentation products may precipitate, so that they can be re- moved by filtration. Otherwise the liquid may be shaken with successive portions of ether, or chloroform, or acetic ether (DRAGENDORFF). The liquid is now nearly or quite neutralized by the addition of fixed alkali hydrate, and tested, as at first, with ferroso-ferric solution, gelatin solution, cinchona sulphate solution, and Fehling's solution, noting if these results differ from those obtained before boiling. The production of glu- coses may be further investigated, by a fermentation test, with yeast (see under Sugars), and by optical examination as to rota- tory power. Tannins are precipitated by lead acetate, copper acetate, and zinc ammonio chloride, and by the salts of nearly all the non alkali metals. The removal of tannins from solutions may be effected by digesting the liquid with recent ferric hydrate, zinc oxide, copper oxide, or lead oxide, and filtration. Also by maceration with animal membrane or rasped hide ; or by filtra- tion through purified animal charcoal. Gelatin gives better precipitates in solution saturated with ammonium chloride or sodium chloride, and the addition of sulphuric or hydrochloric acid further helps the separation of the precipitates. Separa- tions by acetic ether, and non -precipitation, are given under Gallotannin. 1 See note on p. 466. 4 68 TANNINS. ESTIMATION OF TANNINS and Valuation of Tanning Mater- ials. (1) Method of LOWENTHAL.' Titration by a perman- ganate solution, before and after removal of the tannin by gelatin in solution saturated with sodium chloride and acidi- fied. Both titrations made in presence of much indigo solution, which regulates the oxidation and serves as an indicator. The method employs the following-named solutions : (a) Permanga- nate of potassium solution : 1.333 or (Kathreiner) 1.000 gram of the crystallized salt to 1 liter. (J) Indigo solution: 6 grams pure precipitated indigo, and 50 c.c. concentrated sulphuric acid, per liter. There should be added sufficient of the indigo to re- quire for itself two-thirds of all the permanganate used (Kathrei- ner). (c) The solution of Glue and common salt is made by macerating 25 grams of good transparent glue in cold water, then heating to dissolve, making up to 1 liter, and saturating with good common salt. It should be filtered clear when used. (d) The acidulated solution of Common Salt is a saturated solu- tion with addition of 25 c.c. of sulphuric acid in a liter. In the analysis 20 to 25 grams of bark, or 10 grams of sumach or valonia, are boiled with portions of water until fully exhausted, and the solution, when cold, made up to 1 liter. Of this 10 c.c. are diluted to 800 or 1000 c.c., 25 c.c. of the indigo and acid solution are added, the mixture treated with the permanganate solution, drop by drop from the burette, with constant stirring, until the blue color changes to a clear yellow, showing no green tint, and the number of c.c. of permanganate is noted. Another 25 c.c. of the indigo and acid solution are diluted to the same volume made before, and the titration with permanganate re- peated, when this result is subtracted from that first obtained, to obtain the quantity of permanganate require.d for the 10 c.c. of tannin solution. The tannin, as well as gallic acid, if present, are mainly oxidized before the indigo, and therefore oxidized M877: Zeitsch. anal. Chem., 16, 33; Jour. Chem. Soc., 31, 745. KA- THREINER, 1879: Dingier 1 s polyt. Journ.. 227. 481; Zfitxch. anal. Chem., 18, 112; a report fully sustaining this method, and defending it against a criticism in Mohr's Titrirmethode. H. R. PROCTER, 1877: Chem. News, 36, 58; Jour. Chem. Soc., 32, 807, "the most practical method of tannin analysis yet dis- covered." NEUBAUER, Zeitsch. anal, Chem., 10, 1 (1871), after ah elaborate review of methods, gives preference to this. B. HUNT (1885: Jour. Soc. Chem. Ind., 4, 263) reports at length upon this process, and advances modifications, some of which are given in a foot-note further on. Without these modifica- tions Mr. Hunt finds that a considerable quantity of gallic acid may cause too high a result for tannin. Lowenthal made the first report of titration with per- manganate in presence of indigo in 1860: Jour. /. prakt. Chem., 81, 150. The volumetric use of permanganate was introduced by MONIER: Compt. rend., 46, ESTIMATION OF TANNINS. 469 promptly while the permanganate is concentrated. A portion of 100 c. c. of the tannin solution is now treated with 50 c.c. of the glue and common salt solution, and, after stirring with 100 c.c. of the acidulated solution of common salt, again stirred, set aside several hours, and filtered. 1 The filtrate should be perfectly clear. Of this filtrate 50 c.c. (containing 20 c.c. of the tannin solution) are mixed with 25 c.c. of the indigo solution, and the mixture is titrated with the permanganate solution. Another 25 c.c. of the indigo solution, diluted as in the last trial, and titrated, will give the number of c.c. of permanganate to deduct for reduction by indigo. The remainder will be the number of c.c. of permanganate taken by substances other than tannin in 20 c.c. of the tannin solution. Therefore one- half of this num- ber of c.c. will be the number to deduct for decoloration of the permanganate by substances other than tannin in 10 c.c. of the original tannin solution. In the removal of the tannin, both the gelatin solution and the acid and salt solution must be added in sufficient quantity to give a perfectly clear filtrate. 8 The acid and salt solution must not be brought in contact with the gelatin solution before the latter is fully mixed with the tannin solution. The permanganate is to be added slowly, in a white porcelain dish, giving for reduction as much as four minutes with the original solution, and six minutes with the filtrate. It is better to let the gelatin precipitate stand as much as half an hour be- fore filtering ; if filtered earliei the filtrate will consume more permanganate (HEWITT). In preparing the solution from oak- bark or from galls a few drops of acetic acid may be added for preservative effect, and with each portion of water the material may be boiled ten or fifteen minutes. It must not be forgotten that tannins are instable. Duplicate titrations should be made, and should agree to within 0.1 or 0.2 c.c. of permanganate. *B. HUNT (1885) proceeds as follows: 100 c.c. of the tannin solution is treated (in a flask taken dry) with 50 c.c. of a solution of 2 grams of gelatin in 100 c.c. (freshly filtered)*. The flask is shaken, and 50 c.c. of a saturated solution of common salt (containing 50 c.c. of undiluted sulphuric acid per liter) is added, together with a little kaolin or barium sulphate. The solution filters clear. 2 Lflwenthal finds only a very small and nearly constant reduction of per- manganate by gelatin sol'ution, and ascribes this slight reduction to certain oxidizable substances with the gelatin. He infers that these oxidizable sub- stances are precipitated by tannin before all the gelatin is precipitated. By adding 20 c.c. of the gelatin solution the indigo solution consumed 0.4 c.c. more of permanganate solution. Different kinds of gelatin give only a little difference of results. Kathreiner proposes to deduct one-half of the perman- ganate found to be consumed by the given quantity of gelatin solution (Zeit. anal. Chem., 16, 33, 18, 114). 470 TANNINS. The following schedule of quantities may be changed at dis- cretion : For 10 c.c. tannin solution, with 25 c.c. indigo solution a c.c. For the same again a' c.c. For 25 c.c. indigo solution, diluted as before. . b c c. For oxidizable substances in 20 c.c., tannin sol. a -\- a' 2# m. For 50 c.c. filtrate from 100 c.c. tannin sol., 50 c.c. glue sol., and 100 c.c. acid and salt solution c c.c. For another 50 c.c. of the filtrate c f c.c. For 50 c.c. indigo solution, diluted as last above b' c.c. For oxidizable substances other than tannin in 20 c.c. tannin solution c-4-c' -i, -^t- - a = . For tannin in 20 c.c original solution. ....... m n c.c. So far we have only the permanganate value of the original solution, and, from this, of the material taken to be estimated. The permanganate value serves to compare articles containing the same tannin with each other, as oak-bark with oak bark, galls .with galls, etc. A comparison of oak-bark with galls must be taken with some reservation, as different tannins cannot be as- sumed to act with the same equivalent. The permanganate so- lution may be compared with a standard solution of the purest gallo-tannic acid to be obtained, or with any article of tannin of known value. The conditions of time, temperature, and dilu- tion must be kept constant in all comparisons, both in extracting the material and in titrating the solutions. According to the experiments of NEUBAUER, in 1871, 1 63 grams of crystallized oxalic acid (equivalent to 31.4 grams of absolute potassium per- manganate) correspond to 41.57 grams purified gallo-tanriin.* That is, 10 c.c. decinormal oxalic acid solution reduce as much permanganate as 0.04157 grams of Neubauer's purified tannin. Oser has found that the same quantity of oxygen is required for 1 part of gallo-tannin and for 1.5 parts of oak-bark tannin. Then 10 c.c. decinormal oxalic acid solution correspond to 0.06235 grams oak-bark tannin. These factors serve provisionally, Keu- bauer's for galls, sumach, and myrabolans ; Oser's for oak- bark, 1 Zeitsch. anal. Chem., 10, 3. 2 This is near the ratio of 4(H 2 C 2 4 .2H 2 0) to C 14 Hio0 9 ; 504 to 322; 63 to 40.25. The ratio to Schiff's natural tannin glucoside is that of 8(H 2 C 2 O 4 . 2H 2 0) toC 34 H a8 22 ; 63 to 49.25. ESTIMATION OF TANNINS. 471 valonia, and chestnut. No interference in this estimation (with the specified dilutions) by presence of acetic acid, citric acid, tar- taric acid, malic acid, cane-sugar, dextrin, gum, fat, caffeine, or urea (Cech *). Other agents for removal of the tannin in connec- tion with Lowenthal's process have been tried. SIMANDS (1883) proposes to use the gelatigenous tissue of bones, prepared by digesting bone in dilute hydrochloric acid and washing away the earthy chlorides. The tannin solution is macerated with the prepared tissue until the tannin is removed. NEUBAUEK 2 re- moves tannin by purified animal charcoal, which he finds not to remove pectous substances. Lowenthal originally used chlori- nated lime instead of the permanganate. (2) Method of GERLAND, improved by RICHARDS and PAL- MER/ Volumetric precipitation by potassium antimony tartrate in presence of ammonium acetate. Either acetate or chloride of ammonium causes a much closer precipitation of tannin, and pre- vents precipitation of gallic acid. The standard solution of Tar- trate of Antimony and Potassium contains 6.730 grams of the salt dried to a constant weight at 100 C. in one liter. Of this 1 c.c. corresponds to 0.010 of digallic acid. The solution of Ammonium Acetate was prepared by Richards and Palmer by saturating glacial acetic acid with stronger water of ammonia. The material for analysis is dissolved or exhausted so as to fur- nish a solution of 150 c.c. to 300 c.c. in volume and strong enough to contain 0.3 to 0.9 gram of tannin. The entire solu- tion from the weighed portion of material is now divided into three (or four) aliquot parts. To one division the standard solu- tion of antimony is added from the burette, in probable excess, and to a second division a quantity sure not to be an excess is added. To each liquid the acetate of ammonium solution is added, in proportion of 1 c.c. (of the strong solution just speci- fied) to about 25 c.c. of total liquid. The precipitates are left to settle, and as soon as clear liquids appear a drop is taken from each division and tested on a hot porcelain plate with a drop of l Zeitsch. anal. Chem., 7. 134. 2 1871: Zeitsch. anal. Chem., 10, 1. 3 GERLAND, 1863: Chem. News, 8, 54; Zeitsch anal. Chem., 3, 419. GAUHE (1863: Zeitsch. anal. Chem., 3, 131) reports upon the method, with ob- jection on ground of the difficulty of fixing the end of the reaction, and ad- vises to test a nitrate of the titrated liquid for antimony by zinc and hydro- chloric acid upon platinum foil. ^ RICHARDS and PALMER, 1878: Sill. Am. Jour. Sci. [3] 16, 196, 361 modifications, in the substitution of ammonium acetate instead of chloride, and in testing for excess of antimony. These authors report elementary analyses of the precipitates. 472 TANNINS. solution of sodium thiosulphate. If the antimony have been added in excess the orange precipitate of antimonious sulphide will appear. By continued tests of the second division the point of least excess of antimony capable of recognition is found ap- proximately. This point is then fixed with exactness by tests of the third division (and, if provided for, the fourth division). The loss by taking out test-drops is reduced to a minimum in the final titrations. It is better to carry the titrations to a decided orange tint for excess of antimony, and then subtract 0.5 c.c. from the reading of the antimony solution as a correction for this excess. The c.c. X 0.01 = the grams of tannin counted as digallic acid. Gallic acid does not interfere in this method, owing to the ammonium acetate. Various colors occurring in tanning materials enter into the precipitates, some of them uniting with the antimony instead of the tannin, and therefore appearing as tannin in the results. Two classes of color sub- stances are indicated by the experiments of Kichards and Palmer, one closely allied to quercetin, and both related to tannins and tanning agents. With this method, as with Lowenthal's, true comparisons between different tanning agents, as between oak- bark and hemlock-bark, are not likely to be obtained. The for- mula of the typical precipitate of digallic acid is presented by the authors last named as Sbo(C 14 H 8 O 9 ) 3 .6H 2 O. It therefore demands 2KSbO C^O^S X 323 = 646) to 3C 14 H 10 O 9 (3 X 322 = 966). The authors' analyses of precipitates of pure tannins support the formula very well. ^ (3) HAGER'S method with copper oxide. 1 The addition of oxide of copper to take up the tannin, which is estimated from the increase in weight of the oxide, or (as in Hammer's plan) by the decrease in specific gravity of the solution. The powdered material is extracted first with water and then with alcohol, the concentrated solution treated with alcohol and filtered, the filtrate evaporated to remove all the alcohol, diluted with water, filtered, and the solution made up to a determinate volume, of which the specific gravity may be taken. Eecently ignited oxide of copper, equal to at least five times the weight of the tannin to 1 FLECK, in 1860, used precipitation with acetate of copper and volumetric estimation of the excess of copper in solution, by potassium cyanide. SACKUR, Qerberzeitigung, 31, 32, directed tho ignition of the copper precipitate, and WOLFF, 1862: Zeitsch. anal Ohem , i, 103, from twenty-eight analyses gave 1 to 1.304 as the ratio between ignited copper oxide and tannin in the precipitate formed by acetate of copper. To exclude gallic acid Fleck treated the pre- cipitate with ammonium carbonate solution. Hager, in his " Untersnchun- gen," vol. ii. p. 115, gives the method here presented. ESTIMATION OF TANNINS. 473 be found, is now added, the mixture warmed for an hour, and set aside, with occasional agitation, for a day. The filtrate may now be made up to the determined volume, the specific gravity taken, and the table consulted for percentage of tannin corre- sponding to difference of gravity. The precipitate may be washed clean, dried on the water-bath, and weighed, the increase in weight showing the quantity of tannin. Gallic acid will be included. The method seems open to danger of loss of tannin by decomposition, especially with oak-bark tannin. (4) The method of HAMMER 1 has been much used for com- mercial analyses, but gives untrustworthy results. The water solution of the material, made up to a determinate volume, is macerated with dried rasped hide, in quantity at least five times- as much as the tannin to be found, until the tannin is wholly removed from solution. The filtered liquid is made up to the volume before noted, and the specific gravity is to be taken both before and after the removal of the tannin. Difference in spe- cific gravity -f- 1 = specific gravity for per cent. A table of per- centages of gallotannin is given at p. 477. Pectous substances are absorbed by the rasped hide, a cause of error unless the pec- tous substances are removed by precipitation with alcohol, which is then evaporated. (5) The method of WAGNER 3 gives insufficient results with oak-bark, but is serviceable for various manufactured forms of tannins. It is a volumetric precipitation by an alkaloid. 4.523 grams of good sulphate of cinchonine, with 0.5 gram sulphuric acid and 0.1 gram acetate rosaniline or fuchsine, are dissolved in water to make one liter. Each c.c. of this solution precipitates 0.01 gram tannic acid. One gram of solid material is obtained in clear solution of about 50 c.c. measure. To this the standard solution of cinchonine is added, the color being thrown down in the precipitate. By a quick agitation the precipitate soon set- tles. When the tannic acid is all precipitated, the aniline color appears in solution. One gram having been taken, each c.c. of the volumetric solution indicates 1 per cent, of tannic acid. Gallic acid is not precipitated by cinchonine. CLARK 3 has tried a modification of this method for cases of colored liquids which '1860: Jour.f.praU. Chem., 3, 159. 2 1866: Zeilsch. anal. Chem., 5, 9. As to limits and deficiencies of this method see BRAUN, 1868: Zeitsch. anal. Chem., 7, 139. 3 Contributions from Chem. Lab. of Univ. of Mich.. 1876: Am. Chem 7,44. i 474 TANNINS. obscure the aniline red. It is the use of the standard solution of cinchonine in some excess, filtering, washing sparingly, and ti- trating back in the filtrate with Mayer's potassium mercuric iodide solution. This solution may be compared with the cin- chonine solution, or the factor of 0.0124 gram of cinchonine sulphate for each c.c. of Mayer's solution may be used. Among the many other methods for determination of tannins are those with use of Acetate of Lead as a precipitant, with alco- hol ; 1 bone gelatin solution with alum ; 2 and ferric acetate so- lution with sodium acetate. 3 Upon the adaptation of the several methods of estimation to the several well-known different tan- nins, see GUNTHEK, 1870. 4 For estimation of tannin in leather HAGEK'S method may be employed. For the most part gallic acid is obtained from leather instead of tannic acid. 6 Of distinctly known tannins, or tannic acids, the limits of this work permit only the following to be described. GALLOTANNIN. Kutgall-tannin. Gallusgerbsaure. Chiefly Digallic acid, or Gallic anhydride, P TT O C 6 H 2 (OH) 2 CO 2 H ) ^ 009 ^14^10^9 C 6 Ho(OH) 3 CO . \ (SCHIFF), but containing a portion of glucoside of digallic acid. Gallotannic Acid. The TANNIC ACID of the pharmacopoeias and of commerce. Gallotannin is identified as a tannin by its sensible properties {a, J), its reactions with gelatin, alkaloids, iron salts, and perman- ganate (d) ; identified as gallotannin by its fermentation pro- duct (c and p. 467), its product by heat (#), its color with iron salts, with molybdate, and the total bearing of its qualitative tests (d) ; estimated by the method of Lowenthal, Gerland, or Wagner (pp. 468-73) ; separated from metallic compounds, iron inks, and the fruit acids, by acetic ether (c) or by calcium ace- tate (p. 21) ; removed, along with tannins in general, by metallic oxides, gelatin, hide, or bone-black, as described on p. 467. May be prepared from galls as stated on p. 477. 1 SCHMIDT, 1874: Jour. Chem. Soc., 28, 1183. ALLEN, Chem. News, 29, 169, 189. 2 One of the earliest methods: FEELING, MULLER: Liebig and Kopp's Jahresber., 1853, 683; Ding.polyt Jour., 151. 69; Zeitsch. anal Chem., 5,232. 3 HANDTKE, 1853: Jour, f.prakt. Chem., 58,345 4 Pharm. Zeitsch. f. Russland, 1870. Zeitsch. anal. ChC ' j,9panodun) 9un/iojf j,9panod -itnf) 9unfiojf 9U'i t f uosfiff dunoji 9unfioj\[ i S e. TH *> 8 S g M s s eg a s oi o g S ~ s % S ~ 8 3 9 CO co os os ol ? 2 ~ S g j> eo' 8 S 10 si > 1-1 SB 55 d S 10 -i eo g e 3 S I- OJ 10 T* I! ! S S 10 1-1 co eo co 10 10 eo T-I co' eo oi ~ . So t- eo O eo" Tji oi co os $ d C CO O a-. eo OJ co S S si *: jf : 3 I 4) ) l H 5 H 5 i c ^3 H^ W ^i ^ if! 11! * o G wju s p-5 1^2. i >Q gj * ' frj ^ -3ST5S G - S ** d ,^ c C^TJ aj'g sg^SSS 5!1 111 TEAS OF COMMERCE. 2 2 \ * 507 fioiuy uimpvjv 6uojoo vsouuog i39wi &UOIOQ fiv uowiuoo 6wpo Aouty ftowy DUOJOQ os&uuog poof) SUOJOQ VSOUUOjf MU9(InS fe 5 S S S S ( eg S S in ci o jo S JS fe 8 00 ' ^ ^ S Gwioo 6UOJOQ DSOUUOtf S ^ in o o t> s GO 10 IN eo O T-. 9 '>j; umpuj Q '9j; uvipui 'V9J, umpui 'V9tl UWtpltf 'Z "09J, 'I 'V9J, oo o I s'-i . 3 i ?t 1 * : .S S .S " e e 6 1 ? ft 1 * 508 TEAS OF COMMERCE. 1 898HfOUy /0 'OJf CO CD a "PPY -josuf ysy MM MM CO CO N CO * s yay 9jqn2osuf OtCOCS^ IVT^CO 1 ^ N S ^ A SSSS oi TH oi ui eo t-J oi ei SB S S g y#y 9jqniog co eo eo eo' co* oi co eo' S S 8 88 eo' oi co eo' ysy fDjojj S2 2 2 3 3 S 2 CO )Q 1Q CO ,_j *** cooot-oS Si2co^* So e e5 S oi r-i w ci P S umunj, 06 eo" n< o o T-J co i> 8 ^ S S CO OO T t C$ PQ 2 > oS^eo i2ooooS SOD TH <* CO ^ O CO * O O O * 1O ift ^* t oi o fo co gOVAlXg IVfOJj 1 " i 1 i i s i 800 JO QQ ^r CO "^r fe JT 8 fe '3ovj,)x$f ''// i/^/7 COOt^CO G^OQO(^ 8 i 8S 1 '9J,niswf[ 8 35 5S : B S 8 . JS cS S5 . e t> 06 1 ' : 11 o . : : Averages calculated upon t) lea, unless otherwise st 5 * S ? : B 1 s 1 s ' 1 g 1 j||| | { i 1 III! 1141 CONGOUS. Maximum . . . Minimum Average dried at 100 C TEAS OF COMMERCE. 509 -OJY uouiutoo aioswx uoutuioo poof) 2 8 CO J8 oo 8 - noGttoo 6mu noBuoj 6mu Homy WMpffl 8 ~ Suojoo J_A BUOJOQ mow 'Suojoo mom 8 g co eo" s - Suojoo mom -uvj g 3 g s io ei g d os ^ -(nodvnf) mn ^period -unf) S g 3 3 V9J, UWtpllf 9jj umpuj S rt S3 Sis x~, S '^ 4-5.S S3 gss i a a I !. ,d -fia fl^ Qo-a S O> " "^ . .. TH O> I =1 II! a|! (3 ^ O - " tannin, 100 pa E M ^ i-i . 1 &"& dra t eg Ratio (per cent.) tai 10 minutes in 100 p total tannin in lea 1 5 "3 j t 1! Per c draw wate 5io . TEAS OF COMMERCE. the same conditions. The results cannot be considered absolute, but, as they vary only between narrow limits, they are sufficiently accurate to illustrate the behavior of these various teas when subjected to the customary household method of pouring boiling water upon the leaves and allowing it to draw. The results of this table (III.) give the percentages of " ex- tract," theine, tannin, 1 ash (mineral matter) dissolved, the alka- linity of the ash expressed as potassium oxide, and the ratio (per cent.) of "extract" and tannin to the total amount of these two in the leaf. The percentages are calculated upon the air-dried leaf. A comparison of the results for the five Oolong teas shows the finer grades to have yielded more extract, theine, and ash than the poorer grades. The decline, from the fine to the poor grades of the various teas, in the amount of theine dissolved, is something noteworthy, as showing the fine grades to yield nearly all their theine, while the poorer grades do so only to a limited extent. The percent- ages of tannin are quite irregular. Further, the table shows that there is more mineral matter extracted from the leaf than is indicated by the term " soluble ash " in Table I., the difference being .62 per cent, as an average of fourteen determinations. The ratio of tannin to the "extract," and the ratio of either one to the total tannin and " extract " of the leaf, varies quite uniformly with the value of the tea, the per cent, of tannin fall- ing or rising with the percentage of u extract." See Table IV. It will also be noticed that the Congou teas yielded low per- centages of " extract " and tannin, showing that the time allowed for drawing in these teas should be greater than ten minutes, if a full yield of these constituents is desired. If this is uniformly true of Congou teas, they would certainly be suitable for people to whom the large quantity of tannin of the other varieties is objectionable. The tannin extracted from the best green tea was unusually large, being 16.79 per cent. Both Indian teas show a good yield of " extract," theine, tan- nin, and soluble mineral matter. Although these results are quite satisfactory in showing the difference in the drawing quali- ties of various-priced teas, they are not sufficiently uniform to make the results of an analysis the basis for calculating the price of a tea. It is evident that the essential oil plays a more impor- tant part than any other constituent of the tea in determining its commercial value. 'The percentages of tannin are somewhat greater than would be obtained in using a hard water. TEAS OF COMMERCE. TABLE IV. FINEST FORMOSA OOLONG. theine, and ash dissolved from tea by dis- tilled water and Croton water, by allow- 1 1 | 1 | 1 ing to draw from three minutes to over S 5* Is b o "og g one hour. (One hundred parts of boiling water were poured upon one part of tea.) 1 SI 55 . |1 g g g g V TO us ~ Per cent extract total 25 97 28 37 27 47 30 87 30 25 oq r-c Per cent extract less ash 22 25 24 50 23 85 26 70 26 ">2 29 42 Per cent, tannin 9 755 11 23 10 18 13 46 10 60 14 94 Per cent, theine 1 95 2 65 2 02 2 75 2 82 2 85 Alkalinity of ash as potassic oxide 1 029 1 22 1 076 1 22 1 152 1 28 Per cent ash 3.725 3.805 3.625 4.175 4.125 4.325 Table 1Y. illustrates the difference in the drawing quality of an extra choice Oolong tea when treated either with'distilled or Croton water. It shows that in ten minutes' " drawing " the theine was practically extracted, and that the Croton water ex- tracted less tannin than the distilled water, while there was no noteworthy difference in the percentages of extract and ash when the distilled water and Croton water were allowed to draw for the same length of time. Hard waters dissolve less tannin than soft waters under the same conditions. This will also be noticed in the above table. And Table IY. serves to illustrate the rapidity with which the constituents of the tea leaf are dissolv- ed, and that the choice of the water and the proper length of time for drawing are very important factors in preparing a good cup of tea. Practical conclusions. Though varying widely for different teas, the total soluble (extractive) matter averages about 33 per cent., but the average is considerably lower for the infusion of tea prepared by the ordinary household method. The volatile oil gives the flavor and aroma, the tannin and extractive matter the astringency, strength, and body to the infusion. Theine, being almost tasteless, is not taken into account by " tea-tasters," though, physiologically, the most important constituent of the tea. Besides the above, the appearance of the leaf, as well as the color of the infusion and any peculiar foreign taste or smell imparted to the same, have considerable bearing in the " tea- 512 / THEOBROMINE. taster's " method of valuation. A strict relation between tlie chemical composition of the tea and the commercial value of the same is therefore scarcely to be looked for, although the former would disclose at once that tea which is physiologically the best. The principal constituents of tea are the volatile oil, theine, tannin, albuminous compounds, gum, etc., and the soluble mineral matter, containing considerable potash and phosphoric acid. The fertility of the soil, the nature of the climate, the pro- cessing and manipulation the leaves undergo after being pluck- ed, and the care with which the tea is handled thereafter are all instrumental in influencing the chemical composition and the quality of the tea. Uniformity in composition cannot be ex- pected. The principal difference between Green, Oolong, and Congou teas is caused by the processing and manipulation ; but, whatever the modus operandi of the latter, it cannot make good tea out of leaves which have not had the proper conditions of soil and climate to further the production of those constituents which are characteristic of tea. In the ordinary analysis of the tea only the more important constituents are determined, in order to establish the presence or absence of foreign matter. The results thus obtained are scarcely applicable to the commer- cial valuation of tea, since much is there determined which does not enter the infusion of tea. It is the quality of the infusion which is of importance to the consumer, and not the total com- position or appearance of the leaf. Tea is essentially something for the epicurean. To discriminate between qualities of teas of nearly the same grade requires a delicate and sensitive palate. Expert tea-tasters are guided chiefly by the strength, flavor, aroma, and quality of the infusion in judging and classifying tea as to its quality. THALLINE. See CINCHONA ALKALOIDS, p. 168. THEBAINE. See OPIUM ALKALOIDS, p. 358. THEINE. See CAFFEINE, p. 77. THEOSROMINE.-C 7 H 8 N 4 2 = 180. A dimethyl xan- thine, C 5 H 2 (CH 3 ) 2 N 4 O 2 . See Caffeine, p. 77. Found, without caffeine, 1 in the seed of the Theobroma Cacao, or " chocolate 1 SCHMIDT (1883) found a little caffeine in cacao. THEOBROMINE. 513 nut " (WOSKRESENSKY, 1841), and, as a smaller accompaniment of caffeine, in the seed of the Sterculia acuminata, the " cola nut." The dry cacao seed freed from husk, the " cocoa nibs," contains about 1.5 percent, of theobromine (WOLFRAM, 1879) ; while the husks, the '' cocoa shells," furnish from 0.3 to 0.7 per cent, in average yield (WOLFRAM, DONKER, 1880). a. Theobromine crystallizes in the trimetric system, appear- ing in permanent, anhydrous white needles and club-shaped groups, to the unaided eye as a crystalline powder. Sublimes without decomposition, yielding distinct microscopic crystals of sublimate at 170 C. (BLYTH, 1878). Sublimes at 290 to 295 C. (KELLER, 1854). b. Theobromine has a very bitter taste, slowly produced. Its physiological effects are like those of caffeine, but are ob- tained by smaller doses (MITSCHERLICH, 1859). It is excreted in the urine. c. Theobromine is slightly soluble in water or alcohol, its solution requiring 1600 parts water at 17 C. (62.6 F.), and 148 parts water at 100 C. (DRAGENDORFF) ; 4284 parts absolute alco- hol at 17 C., and 422 parts boiling absolute alcohol (TREUMANN, 1878), in 1400 parts cold alcohol (MITSCHERLICH, 1 859). It is but very slightly soluble in ether, one part requiring 17000 parts cold ether or 600 parts boiling ether (Mitscherlich). It dissolves in 105 parts boiling chloroform (Treumann) ; is somewhat solu- ble in amyl alcohol ; but slightly soluble in benzene ; insoluble in petroleum benzin. Theobromine is a weak base. It forms crystallizable salts ; but on contact with water they give up acid and become basic salts, and those of volatile acids give up the acid at or below 100 C. Theobromine dissolves in hydrochloric and in other acids ; but the hydrochloride, C 7 H 8 N 4 O 2 . HC1 . H 2 O, and the nitrate, C 7 H 8 N 4 O 2 .HNO 3 , do not dissolve at all freely in water alone without free acid. Theobromine dissolves in ammonia-water. Respecting combinations, see report of Messrs. SCHMIDT and PRESSLER, 1883. ' d. Theobromine responds to the murexoin test with the same intensity as Caffeine (p. 79), forming amalic acid when warmed with hydrochloric acid and potassium chlorate and evaporated to dryness on the water-bath, and giving purple- colored murexoin when the cold residue is touched with am- monia. Phosphomolybdate of sodium, added to the acidulous 1 Liebig's Annalen, 217, 287; Jour. Chem. Soc., 44, 872. TYRO TO XI CON. solutions of theobrornine, gives a yellow precipitate, obtained in dilute solutions. Platinum chloride does not precipitate, except in concentrated solutions, when crystals are obtained, (C 7 H 8 N 4 O 2 ) 2 H01PtCl e .4H 2 O. In like manner gold chloride yields yellow crystals, in tufts of needles, C 7 H 8 N 4 O 2 . HC1 . Au01 3 . When an ammonia solution of theobromine is treated with silver nitrate solution, a gelatinous precipitate is obtained, and on boiling this granular crystals of argentic theobromine are ob- tained, C 7 H 7 AgN 4 O 2 . And when this compound is treated with anhydrous methyl iodide, at 100 C., for twenty-four hours, caffeine (methyl theobromine) is formed, with silver iodide (STRECKER, 1861). Again, when theobromine, alcoholic solution of potassium hydrate, and methyl iodide, in equiva- lent quantities, are heated together at 100 C. in sealed tubes, caffeine is formed, with potassium iodide (SCHMIDT and PRESSLER, 1883). C 7 H 7 Ag]Sr 4 2 + CH 3 I = C 7 H 7 (CH 3 )ISr 4 2 + Agl C 7 H 8 N 4 O 2 + CH 3 I + KOH = C 7 H 7 (CH 3 )N 4 O 2 + KI + H 2 O. Potassium mercuric iodide produces no precipitate in the acidu- lous solutions of theobromine, and iodine in potassium iodide solution causes little precipitation (distinctions of caffeine and theobromine from most other alkaloids). e. Theobromine may be separated from non-volatile mat- ters, in general, by sublimation at a gradually increasing heat beginning at 170 C. From most alkaloids by its slight solubi- lities, and from caffeine by its smaller solubilities in benzene (SCHMIDT), or water, or ether. > The quantitative estimation of theobromine in cacao is made by SCHMIDT and PRESSLER (1883) as follows : The crushed cacao is freed from oil by pressure, half its remaining weight of slaked lime is added, and the mixture is boiled repeatedly with alcohol of 80 per cent, strength. The residue on evapora- tion of the alcohol is recrystallized from the same solvent, and is obtained as a white, crystalline powder. It may be dried at 100 C. and weighed. TROPEINES. See MIDRIATIC ALKALOIDS, p. 339. TURKEY-RED OIL. See FATS AND OILS, p. 287. TYROTOXICON. " Cheese Poison." The putrefactive product obtained in 1885 by Professor Yaughan, and recently TYROTOX1CON. 515 announced by him to be diazoberizene, C 6 !I 5 .N:N, in combina- tion with acids. 1 a. Tyrotoxicon, obtained from milk products as direct- ed under e, was found to agree with diazobenzene butyrate, C 6 H 5 . E~ 2 . C 4 H 7 O 2 , in crystallizing in needles, which gradually decompose in moist air. Potassium diazobenzene, C 6 H 5 . N 2 . OK, obtained from tyrotoxicon, 2 appeared in line six sided plates. Tyrotoxicon compounds, at 100 C., explode with violence. b. The crystals have a penetrating, old-cheesy odor. A minute portion placed upon the tongue produces " dryness of the throat, nausea, vomiting, and diarrhoea." In children the effects agree with the symptoms of cholera infantum. Ten drops of a concentrated aqueous solution of tyrotoxicon from milk three months old, placed in the mouth of a small dog three weeks old, in a few minutes caused " frothing at the mouth, retching, vomiting of frothy liquid, rapid breathing, muscular spasm over the abdomen, and after some time watery stools." Similar ef- fects were obtained with cats, and subsequent dissection showed the mucous membrane of the stomach and intestines to be blanched and soft. Of diazobenzene butyrate, artificially pre- pared, 3 0.010 to 0.025 gram given to cats caused severe symptoms, the same as above detailed, and 0.100 gram caused death, the mu- cous membrane of the stomach not being reddened, but left pale and soft. G. " Tyrotoxicon is soluble in water, alcohol, chloroform, and ether." " Purified tyrotoxicon is insoluble in ether, and it probably owes its solubility in ether at this stage to the presence of impurities." The ordinary salts of diazobenzene are more or less freely soluble in water, sparingly soluble in alcohol, and are for the most part precipitated from alcoholic solutions by ether. d. The diazobenzenoid compounds are identified by the reaction of LIEBEKMANN,* namely, by the bright colors they give 'VICTOR C. VAUGHAN, 1884-85: "A Ptomaine from Poisonous Cheese," Zeitsch. phyxiolog. Chem., 10, 146; Jour. Chem. Soc., 50, 373. Michigan State Board oi' Health Reports, 1885 and after. "Tyrotoxicon: Its presence in poi- sonous cheese, ice-cream, and milk," Am. Assoc. Adv. Sci., Buffalo Meeting, August, 1886, Jou/: A'nalyt. Chem., i, 24. "The Chemistry of Tyrotoxicou and its action upon the lower animals," with report of determination of diazo- benzeno, ibid., i, 281. * By method of GRIESS, 1866: Ann. Chem. Phar., 137, 54. 3 By the method of (JRIKSS, loc. cit. 4 LiEBERMANN, 1874: Bar. d. chem. Ges. t 7,247; Jour Chem. Soc., 27. 693: that sulphuric acid holding nitrous acid in solution gives color-reactions 5 1 6 TYRO TOXICON. when treated with concentrated sulphuric acid and phenol. " "With equal parts of sulphuric acid and carbolic acid the pre- pared [artificial] diazobenzene nitrate gave a green coloration ; while with the same reagents tyrotoxicon gave a color which varied from a yellow to an orange-red. But the diazobenzene nitrate dissolved in the whey of normal milk and extracted with ether, or in the presence of other proteids, gave the same shades of color as the tyrotoxicon did, and the potassium compound of tyrotoxicon prepared by the method to be given later produced the same shade of green as did the artificial diazobenzene. This color test may be used as a preliminary test in examining milk for tyrotoxicon. It is best carried out as follows : Place on a clean porcelain surface two or three drops each of pure sulphuric acid and pure carbolic acid. This mixture should remain color- less, or nearly so. Then add a few drops of the residue left after the spontaneous evaporation of the ether. If tyrotoxicon be present a yellow to an orange-red color will be produced. This test is to be regarded as a preliminary one ; for it may be due to the presence of a nitrate or nitrite. 1 The tyrotoxicon must be purified according to a method to be given further on before the absence of nitrate or nitrite can be positively demonstrated." The explosion of tyrotoxicon may be obtained, in evidence of its identity, by exposure of the platinochloride to a tempera- ture approaching 100 C., as in the discovery of this property by Prof. Yaughan. A solution of the tyrotoxicon in absolute alco- hol is treated with a little platinum chloride, and heated in an open dish upon the water- bath, when, as the alcohol is nearly or quite all vaporized, the explosion results. The known diazo pla- tinum compound is (C 6 H 5 . N 2 . Cl) 2 PtCl 4 , and in explosion is resolved into 2C 6 H 5 C1 + N 2 + 2C1 2 + Pt. The aurochloride of tyrotoxicon is obtained, in precipitate or in golden plates, as follows : " In the nitrate from milk which is rich in tyrotoxicon, after neutralization with sodium carbonate, with phenols generally, the produced colors containing nitrogen but not in the form of the nitro, and probably not in that of the nitroso group. All 'diazo compounds, also the diazo-amido compounds, share with the ni- troso compounds (Hofmann) and the nitrites (Liebermann) the power to give with sulphuric acid and phenol red to blue colors of the utmost intensity (E. FISCHER, 1875). The color-products so obtained are azo-benzenoid bodies, well known as azo dyes, represented by the tropceolines (this work, p. 186; O. N. WITT, Jour. Ghem. Soc., 35, 179). The azo compounds, it will be remembered, contain the group NN interposed between two benzenoid (or other carbona- ceous) groups. Thus, C 6 H 4 .N 2 . S0 3 (diazobenzene sulphonic acid)+0 6 H 5 OH = CeH-t.SOaH.Na.CeBU.OH (oxy-azo-benzene sulphonic acid). 1 " The coloration with nitrates and nitrites is darker than with diazoben- TYROTOXICON. 517 filtration and acidifying with hydrochloric acid, gold chloride produces a precipitate which is insoluble in water, but soluble in Aot alcohol, from which it separates, on cooling, in golden plates." " The gold compound is decomposed by frequent treatment with hot alcohol." The potassium compound of tyrotoxicon believed to be po- tassium diazobenzene, C 6 H 5 .N 2 .OK was prepared as follows: " The aqueous residue [see e] was acidified with nitric acid, then treated with an equal volume of potassium hydrate and the whole concentrated on the water-bath. . . . On cooling the mass crystallized ... in six-sided plates, along with the prisms of po- tassium nitrate. The crystalline mass obtained from the tyro- toxicon was treated with absolute alcohol, filtered, the filtrate evaporated on the water-bath, the residue dissolved in absolute alcohol, from which it was precipitated in a white, crystalline form with ether." The decompositions of tyrotoxicon are so far found by its discoverer to agree with the well-known decompositions of diazo- benzene salts. Warmed with water the latter break up into car- bolic acid and nitrogen, thus : C 6 H 5 . N 2 . M) 3 + H 2 = C 6 H 5 . OH + ]ST 2 Warmed with alcohol, aldehyde and hydrocarbons result as fol- lows : C 6 H 5 . N 2 . K0 3 + C 2 H 6 = C 6 H 6 + N 2 + C 2 H 4 With the acids of the halogens changes occur as follows : C 6 H 5 . 1ST 2 . ]ST0 3 + HC1 = C 6 H 5 C1 + N 2 + HNO 3 . The reducing agents in general cause immediate decomposition. Hydrogen sulphide reacts promptly. e. The following directions for the separation of tyrotoxicon from milk or cheese are taken from the last article of Dr. Vaughan : " Milk or other fluid to be tested for this poison should be kept in well-stoppered bottles ; for if the fluid be ex- posed to the air the tyrotoxicon may decompose in a few hours. The filtrate from the milk, or the filtered aqueous ex- tract of cheese, should be neutralized with sodium carbonate, then shaken with half its volume of pure ether. Time should be given for the complete separation of the ether. . . . After complete separation the ether should be removed with a pipette and allowed to evaporate spontaneously in an open dish. The residue from the ether may be dissolved in distilled water and again extracted with ether ; but repeated extractions with ether are to be avoided, for as the tyrotoxicon becomes purified it be- 5i8 VALERIC ACIDS. comes less soluble in ether. To a drop of an aqueous solution of the ether residue apply the preliminary test with sulphuric and carbolic acids. To the remainder of the aqueous solution of the ether residue add an equal volume of a saturated solution of caustic potash, and evaporate the mixture on the water-bath. The double hydrate of potassium and diazobenzene[C 6 H 5 .N 2 .OK] will be formed if tyrotoxicon be present." The recognition of potassium diazobenzene is stated on page 517. f. An estimation of tyrotoxicon is indicated by the experi- ments of Vaughan, in converting the potassium compound (d), prepared as directed (#), into potassium sulphate for weight. The white, crystalline precipitate, by the ether, " was collected, washed with ether, dried, and the per cent, of potassium esti- mated as potassium sulphate.'' 1 K 2 S0 4 : 2C 6 H 5 N 2 OK : 2C 6 H 5 N 2 lSrO 3 ::174 : 320 : 334. ULTIMATE ANALYSIS OF CARBON COMPOUNDS See p. 201. VALERIC ACIDS. C 5 H 19 O 2 == 102 (monobasic): Pri- mary Pentoic Acids. Four pentoic acids are theoretically pos- sible, as oxidation products of the four primary pentoic alcohols, and are all known, as follows : (1) Normal valeric acid, CI1 3 .CII 2 .CH 2 .CH 2 .CO 2 H. Made from normal butyl cyanide, etc. (2) Isovaleric acid. Inactive valeric acid. (CH 3 ) 2 CH.CH 2 . CO 2 H. Isobutyl-carboxyl. Chief valeric acid of vale- rian oil. Obtained by oxidation of the chief alcohol of fusel oil. (3) Methyl-ethyl acetic acid. Active (dextro-rotatory) valeric acid. CH 3 .C 2 H 5 .CH.CO 2 H. In small proportion in valerian oil, according to some observers. Obtained by oxidation of the lesser pentyl alcohol (13$) of fusel oil. (4) Methyl-prcpyl acetic acid, CII 3 C 3 H 6 .CO 2 H. Made from methy 1-propy 1- carbin ol . ORDINARY VALERIC ACID. ISOVALERIC ACID Inactive Valeric Acid. The second of the pentoic acids above named. Baldriansaure. A constituent of u valerian root," the rhizome and rootlets of Valeriana officinalis, and a part of the volatile oil of valerian. Reported as found in digitalis, Artirnisia Absin- 1 Per cent, of potassium calculated, 24.42; found, 23.92. ORDINA RY VA LERIC A CID. 5 1 9 thium, Anthemis nobilis, Sambucus nigra, Yiburnum opulus, and other plants. Manufactured by oxidation (distillation from dichromate and sulphuric acid) of isoamyl alcohol (isobutyl car- binol), the principal alcohol of fusel oil. Valeric acid is recognized by its odor and the odor of amyl valerate (&), its solubilities ( Dimethylprotocatechuic acid 18 Dimethyl xanthine 212 Diphenylamine yellow . . .<> 186 Diquinicine 91, 95 Dragendorff's plan for plant analysis 423 Dragendorffs process for alka- loids 33 Drying oils 281 Duboisia, alkaloids of 340 Duboisine 340 Ecgonine 170, 172 Eisessig 8 Elaidic acid. . ..,47 Elaidin test 281 Elateriri in color-tests with Froehde's reagent 51 Elaterin in color- tests with sul- phuric acid 50 Elaterin, in plant analysis 425 Elementary analysis 198 Elementary analysis, inorganic and organic 392 Elementary organic analysis, quantitative 201 Eosins 183 Bosin scarlet 183 Bricolin, in plant analysis 425 Erlenmeyer's furnace 208 Erythroxylon Coca 170 Essigsaure 7 Essigsauren Kalk 11 Ethyl-orange 186 Extraction apparatus 409 Extraction-apparatuses for liquids (illustrated) 38 Extract of belladonna, assay of. . . 353 Extract of nux-vomica, assay of.. 457 Fat acids, percentages of, insol- uble 256 Fat acids, quantitative determina- tions of 250 Fat oils, specific gravity of 262 Fats and Oils 238 Fatty acid series 239, 245, 246 Filter of Gooch 409 Flavanilin 185 Fleischer's estimation of tartaric acid 495 Formic acid 312 Formulas, deduction of 237 Froehde's reagent for alkaloids ... 51 Fruits, percentage of citric acid in 85 Fusel oil 314 Fustic color 193 Fustic tannin 479 Gadinine.. . 427 5 26 INDEX. Gallein 183 Gallic acid 320 Gallic anhydride 474 Gallo-cyanin 188 Q-allotannin 474 Gaseous bodies, organic combus- tions of 216 Gaulthoria, oil'of 433 Geisler's report on teas 505 Gelseminine in color-test with sul- phuric acid . 50 Gelseminine in color-test with ni- tric acid 52, 454 Gelsemine, in plant analysis 425 Gerbsauren 465 Gerland's method for estimating tannins 471 Gerrard's test for atropine 348 Glacial acetic acid 8, 14 Glaser's combustion furnace 216 Glucose, in plant analysis. . .415, 419, 425 Glucosides, in plant analysis.. 413, 419, 424 Glucoside-tannins 466 Glycerides, as a chemical class. . . 238 Glycerin 323 Glycerin, tests of purity of 328 Gnoscopine 360 Gnoscopine in color-test with sul- phuric acid 50 Gold chloride with alkaloids 49 Gooch's filter 409 Gratiolin, in plant analysis 425 Green Coloring Matters 186, 193 Green oil or anthracine oil 395 Guarana, assay of 81 Guaranine 77 Gums, in plant analysis 416, 420 Hager's method for estimating tannins 472 Halogens, estimation of 236 Hammer's method for estimating tannins 473 Hard pitch 395 Hectographic ink 483 Hehuer's method for fats 250 Hehner's number, interpretation of 301 Helleborin, in plant analysis 425 Hemepic acid 362 Hemlock bark, tannin of 481 Hempseed oil, drying test of 282 Helvetia green 187 Herapathite 131 Hesse's test for quinine sulphate.. 151 Hippuric acid in urine 62 Hippuric acid, source of benzoic.. 60 Hof mann's violet 188 Holzessigsauren Kalk 11 Homatropine 343 Homocinchonidine 93 Homoquinine 92 Hop bitter, in plant analysis. , . . 425 Hop-tannin 481 Hiibl's method with fats 258 Humus, in plant analysis 417, 420 Hydrastine 329 ' ' Hydrastine," yellow alkaloid ... 72 Hydrastis, assay of 74 Hydrastis, constituent of 72 Hydrocinchonidine 91 Hydrocinchonine 93 Hydroconquinine 93 Hydrocotarnine 360, 362 Hydrocyanic acid, from amygda- lin 57 Hydrogen, estimation of 208 Hydrogen, qualitative analysis for 198 Hydroquinidine 91 Hydroquinine 91 Hydroxy-benzoic acids 433, 443 Hydroxy-xylenic acids 443 Hygrine 170, 173 Hyoscine 342 Hyoscyamine 34 Hyoscyamus, alkaloids of 340 Hyoscyamus assay 353 Hyoscyamus leaves and seeds, assay of 353 INDEX. 527 Hypogaic acid 246, 249 Igasurine 446 Immiscible solvents 33 Inactive valeric acid 518 Indelible inks 483 India-ink 482 Indigo-blue 192 Indigo-carmine 188 Induline R 188 Inks 482 Ink-stains, discharge of 484 Inorganic analysis, relations to organic 393 Inorganic substances, in organic analysis 200 Inulin, in plant analysis 425 Iodine and methyl green 1$7 Iodine,, estimation of 236 Iodine numbers of fats 258 Iodine, qualitative analysis for. . . 200 Iodine reactions with alkaloids ... 42 lodophenin 187 Iron-bluing tannins 466 Iron-greening tannins 466 Isobutyl-carboxyl 518 Isobutyric acid (foot-note) 75 Isovalerates(isovalerianates) 519 Isovaleric (isovalerianic) acid 518 Itaconic acid . 31 Japaconine 18 Japaconitine - 18 Jaune N 186 Jervine in color-test with sulphu- ric acid ... 50 Johnson and Jenkins's method . . . 220 Kairines 167 Kerner's test for quinine, 139, 144, 146 Kjeldahl's method for nitrogen . . . 234 Koffein 77 Kottstorfer's method for fats 254 Kottstorfer's number, interpreta- tion of 304 Lanthopine 360, 362 Lard 290 Lard oil 292- Lard, tests of purity of 291 Laudanine 360, 362 Laudanosine 360, 362 Laudanum assay 385 Laurie acid 245 Laut's violet. 188 Lead chromate, for organic analy- sis 203 Lees of tartar 496 Lemon-juice, assay of 89 Leucoline 165 Leucomaines 428 Leukindophenol 188 Lichen-red 192 Liebig's test for quinine 151 Light oil, of coal-tar distillations. . 395 Lignose, in plant analysis. . .418, 420, 426 Lime-juice 85, 89 Linoleic acid 249 Linoxyn 249 Linseed oil 284 Linseed oil, tests of purity of 284 Liquids, organic combustions of.. 213 Liver, excretion of aconite alka- loids in 29 Liver, excretion of morphine in. . 372 Liver, excretion of strychnine in. . 450 Lobeline, in plant analysis 425 Logan in 447 Logwood blue 192 Logwood in inks 482 Lowenthal's method of estimating tannins 468 Luteolin 186 Mace oil, melting of 269, 272, 274 Madder colors 189 Madder-red 193 Madder-violet 194 Magdala-red 182 Magenta 183 Malachite green. . . , 187 Malic acid. . . 333 5 28 INDEX. Manchester brown 186 Margaric acid 244 Martin's yellow 185 Mate, assay of , 81 Mauvein 188 Mayer's solution 43 Mean molecular weight of fat acids 261 Meconic acid 337 Meconic acid, as analytical proof of opium 370 Meconidine 360 Meconidine in color-test with sul- phuric acid 50 Meconin 362 Meissl's method for fats 253 Melting and congealing points of fats 265 Menyanthin, in plant analysis . . . 425 Metacresol 394 Metatungstic acid with alkaloids 43 Metaxylenols 394 Methylene blue 187 Methyl-orange 186 Methyl-theobromine 77 Microscopical characteristics of alkaloids 53 Microscopical distinctions of cin- chona alkaloids 101 Microsublimation of alkaloids 53 Middle oil, in coal-tar distillation 395 Midriatic alkaloids 339 Milk, examination of, for salicylic acid 440 Mineral oils, separation from gly- cerides. 274 Molecules as final products of chemism 391 Morintannic acid 479 Morintannin 479 Morphine 362 Morphine in color-test with sul- phuric acid 50 Morphine in color-test with Froehde's reagent 51 Morphine, salts of 364, 365 Morphine, tests of purity of 386 Moms tinctoria 479 Murexid 80 M urexoin 80 Murexoin test for theobromine . . . 513 Muscarine 427 Mygdalein 427 Myristic acid 245 Mytilotoxine 428 Narceine 359, 362 Narceme in color-test with sul- phuric acid 50 Narceine in color-test with nitric acid.. 52 Narcotine 387 Narcoline in color-test with sul- phuric acid 50 Narcotine in color-test with nitric acid.. ' 52 Neuridine 427 Neurine ..427,428 Nitrogen and carbon, relative de- termination ! . .. 233 Nitrogen estimation 220, 229, 230, 233 ; 234 Nitrogen, total, in plant analysis . . 411 Nitrosalicylic acids 438 Nutgails in inks 482 Nutgalls, treated for preparation of tannic acid 477 Nutgall-tannin ... 474 Nutmeg oil, melting of 269, 272 Nux-vomica, alkaloids 447 Nux-vomica, assay of 456 Oak-bark tannin 478 Oils and fats 233 Oil of birch 433 Oils, fixed 238 Oils, fixed, in plant analysis. ..418. 424 Oils, volatile, in plant analysis. . .412, 423 Olive oil, tests of purity of 285 Opianic acid 362, 388 Opium alkaloids 358 Opium assay ^ 374 INDEX. 529 Orcein 192 Organic analysis, divisions of. ... 391 Organic matter 391 Orseille 192 Oxalate of calcium, in plant analy- sis 426 Oxygen, direct estimation of. ... 234 Oxymorphine 359 Naphthalene-Carmine 182 Naphthalene, source of benzoic acid 60 Naphthalene-Yellow 185 Nataloin 54 Nitric acid reactions with alka- loids 51 Nitrogen, qualitative analysis for.. 199 Okie acid 246 Olein 246 Oleomargarin 292 Olive-kernel oil 287 Olive oil 285 Opium alkaloids 358 Opium, assay of 374 Orange II 186 Orange G 186 Oranges, constituent of 85 Otto's process for alkaloids 33 Oxalic acid, separation from fruit juices 336 Oxygen gas, for organic combus- tions 202 Oxylinoleic acid 249 Palmitic acid 244 Palm oil, melting and congealing of.. 269, 274 Papaverine 359, 362 Papaverine in color-test with sul- phuric acid 50 Papaver somniferum 358 Paracresol 394 Paraffin, melting and congealing. . 272 Paraffins, separation from glyce- rides 274 Paratartaric acid.. .432 Paraxylenol 394 Paricine 93 Par sons' s plan for plant analysis. . 408 Pathological tannins 467 Paulinia, constituent of 77 Paytamine 93 Paytine 92 Peanut oil, melting of 269, 274 Pectous substances, in plant analy- sis 416, 420, 421, 425 Pe 7 cloth, gilt tops $4.00 PYNCHON, Prof. T. R. Introduction to Chemical Physics, designed for the use of Academies, Colleges, and High Schools. Illustrated with numerous engravings, and containing copious experiments with directions for preparing them. New edition, re- vised and enlarged, and illustrated by 269 illustrations on wood. 8vo, cloth $3.00 24 D. VAN NOSTRAND COMPANY'S RADFORD, Lieut. CYRUS S. Handbook on Naval Gunnery. Prepared by Authority of the Navy Department. For the use of U. S. Navy, .U. S. Marine Corps and U. S. Naval Reserves. Revised and enlarged, with the assistance of Stokely Morgan, Lieut. U. S. N. Third edition. 12mo, flexible leather $2.00 RAFTER, GEO. W. and M. N. BAKER. Sewage Dis- Disposal in the United States. Illustrations and folding plates. Second edition. 8vo, cloth $6.00 RAM, GILBERT S. The Incandescent Lamp and its Manufacture. 8vo., cloth $3.00 RANDALL, J. E. A Practical Treatise on the Incan- descent Lamp. Illustrated. 16mo, cloth 50 RANDALL, P. M. Quartz Operator's Hand-book. New edition, revised and enlarged, fully illustrated. 12mo, cloth $2.00 RANKINE, W. J. MACQUORN. Applied Mechanics. Comprising the Principles of Statics and Cinematics, and Theory of Structures, Mechanism, and Machines. With numerous diagrams. Fifteenth edition. Thoroughly revised by W. J. Millar. 8vo, cloth $5.00 - Civil Engineering. Comprising Engineering Sur- veys, Earthwork, Foundations, Masonry, Carpentry, Metal- Work, Roads, Railways, Canals, Rivers, Water- Works, Harbors, etc. With numerous tables and illustrations. Twentieth edition. Thoroughly revised by W. J. Millar. 8vo, cloth $6.50 Machinery and Mill work. Comprising the Geom- etry, Motions, Work, Strength, Construction, and Objects of Machines, etc. Illustrated with nearly 300 woodcuts. Seventh edi- tion. Thoroughly revised by W. J. Millar. 8vo, cloth $5.00 The Steam-Engine and Other Prime Movers. With diagram of the Mechanical Properties of Steam, folding plates, numerous tables and illustrations. Thirteenth edition. Thoroughly revised by W. J. Millar. 8vo, cloth $5.00 Useful Rules and Tables for Engineers and Others. With appendix, tables, tests, and formulae for the use of Electrical Engineers. Comprising Submarine Electrical Engineering, Electric Lighting, and Transmission of Power. By Andrew Jamieson, C.E., F.R.S.E. Seventh edition. Thoroughly revised by W. J. Millar. Crown 8vo, cloth $4.00 A Mechanical Text Book. By Prof. Macquorn Rankine and E. F.B amber, C. E. With numerous illustrations. Fourth edition. 8vo, cloth $3.50 RAPHAEL, F. C. Localisation of Faults in Electric Light Mains. 8vo, cloth $2.00 RECKENZAUN, A. Electric Traction on Railways and Tramways. 213 Illustrations, 12mo, cloth $4.00 SCIENTIFIC PUBLICATIONS. 25 REED'S ENGINEERS' HAND-BOOK, to the Local Marine Board Examinations for Certificates of Competency as First and Second Class Engineers. By W. H. Thorn. With the answers to the Elementary Questions. Illustrated by 297 diagrams and 36 large plates. Sixteenth edition, revised and enlarged. 8vo, cloth. .$5.00 REED'S Key to the Sixteenth Edition of Reed's En- gineer's Hand-book to the Board of Trade Examinations for First and Second Class Engineers and containing the working of all the ques- tions given in the examination papers. By "W. H. Thorn. 8vo, cloth $3.00 Useful Hints to Sea-going Engineers, and How to Eepair and Avoid "Break Downs;" also Appendices Containing Boiler Explosions, Useful Formulae, etc. With 36 diagrams and 4 plates. Second edition, revised and enlarged. 12mo, cloth. . .$1.40 Marine Boilers, A Treatise on the Causes and Pre- vention of their Priming, with Remarks on their General Manage- ment. Illustrated. 12mo, cloth $2.00 REINHARDT, CHAS. W. Lettering for Draftsmen, Engineers and Students. A Practical System of Free-hand Lettering for Working Drawings. Fourth thousand. Oblong, boards. . .$1.00 RICE, J. M., and JOHNSON, W. W. On a New Method of obtaining the Differential of Functions, with especial reference to the Newtonian Conception of Bates or Velocities. 12mo, paper 50 RINGWALT, J. Li. Development of Transportation Systems in the United States, Comprising a Comprehensive Descrip- tion of the leading features of advancement from the colonial era to the present time, in water channels, roads, turnpikes, canals, railways, vessels, vehicles, cars and locomotives ; the cost of transportation a various periods and places by the different methods ; the financial engineering, mechanical, governmental and popular questions that have arisen, and notable incidents in railway history, construction and operation. With illustrations of hundreds of typical objects. Quarto, half morocco $7.50 RIPPER, WILLIAM. A Course of Instruction in Machine Drawing and Design for Technical Schools and Engineer Students. With 52 plates and numerous explanatory engravings. Folio, cloth $6.00 ROEBLING, J. A. Long and Short Span Railway Bridges. Illustrated with large copperplate engravings of plans and views. Imperial folio, cloth $25.00 ROGERS, Prof. H. D. The Geology of Pennsylvania. A Government Survey, with a General View of the Geology of the United States, essays on the Coal Formation and its Fossils, and a description of the Coal Fields of North America and Great Britain. Illustrated with plates and engravings in the text. 3 vols, 4to, cloth, with portfolio of maps $15.00 26 D. VAN NOSTRAND COMPANY'S ROSE, JOSHUA, M.E. The Pattern-Makers' Assistant. Embracing Lathe Work, Branch Work, Core Work, Sweep Work, and Practical Gear Constructions, the Preparation and Use of Tools, together with a large collection of useful and valuable Tables. Eighteenth edition. Illustrated with 250 engravings. 8vo, cloth $2.50 - Key to Engines and Engine-running. A Practical Treatise upon the Management of Steam Engines and Boilers for the Use of Those who Desire to Pass an Examination to Take Charge of an Engine or Boiler. With numerous illustrations, and Instructions Upon Engineers' Calculations, Indicators, Diagrams, Engine Adjustments, and other Valuable Information necessary for Engineers and Firemen. 12mo, cl $2.50 SABINE, ROBERT. History and Progress of the Electric Telegraph. With descriptions of some of the apparatus. Second edition, with additions. 12mo, cloth $1.25 SAELTZER, ALEX. Treatise on Acoustics in connec- tion with Ventilation. 12mo, cloth $1.00 SALOMONS, Sir DAVID, M.A. Electric-Light Instal- lations. A Practical Handbook. Eighth edition, revised and en- larged with numerous illustrations. Vol. I., The management of Accumulators. 12mo, cloth $1.50 Vol. II., Apparatus, 296 illustrations. 12mo., cloth $2.25 Vol. III., Applications, 12mo. , cloth $1.50 SANFORD, P. GERALD. Nitro-Explosives. A Prac- tical Treatise concerning the Properties, Manufacture and Analysis of Nitrated Substances, including the Fulminates, Smokeless Pow- ders and Celluloid. 8vo. cloth, 270 pages $3.00 SATJNNIER, CLAUDIUS. Watchmaker's Handbook. A Workshop Companion for those engaged in Watchmaking and allied Mechanical Arts. Translated by J. Tripplin and E. Bigg. Second edition, revised with appendix. 12mo, cloth $3.50 SCHELLEN, Dr. H. Magneto-Electric and Dynamo- Electric Machines : their Construction aad Practical Application to Electric Lighting, and the Transmission of Power. Translated from the third German edition by N. S. Keith and Percy Neymann, Ph.D. With very large additions and notes relating to American Machines, by N. S. Keith. Vol. I., with 353 illustrations. Second edition.. .$5.00 SCHUMANN, F. A Manual of Heating and Ventilation in its Practical Application, for the use of Engineers and Architects. Embracing a series of Tables and Formulae for dimensions of heating, flow and return pipes for steam and hot- water boilers, flues, etc. 1. mo, illustrated, full roan $1.50 - Formulas and Tables for Architects and Engineers in calculating the strains and capacity of structures in Iron and Wood. 12mo, morocco, tucks $1.50 SCIENCE SERIES, The Van Nostrand. [See List, p. 33] SCIENTIFIC PUBLICATIONS. 27 SCRIBNER, J. M. Engineers' and Mechanics' Com- panion. Comprising United States Weights and Measures. Mensura- tion of Superfices and Solids, Tables of Squares and Cubes, Square and Cube Roots, Circumference and Areas of Circles, the Mechanical Powers, Centres of Gravity, Gravitation of Bodies, Pendulums, Spe- cific Gravity of Bodies, Strength, Weight, and Crush of Materials, Water- Wheels, Hydrostatics, Hydraulics, Statics, Centres of Percus- sion and Gyration, Friction Heat, Tables of the Weight of Metals, Scantling, etc., Steam and the Steam-Engine. Twentieth edition, revised. 16mo, full morocco SI. 50 SEATON, A. E. A Manual of Marine Engineering. Comprising the Designing, Construction and Working of Marine Machinery. With numerous tables and illustrations reduced from Working Drawings. Fourteenth edition. Revised throughout, with an additional chapter on Water Tube Boilers. 8vo, cloth. 1899. $6.00 and ROUNTHWAITE, H. M. A Pocketbook of Ma- rine Engineering Rules and Tables. For the use of M irine Engineers and Naval Architects, Designers, Draughtsmen, Superintendents, and all engaged in the design and construction of Marine Machinery, Naval and Mercantile. Fifth edition, revised and enlarged. Pocket size. Leather, with diagrams $3.00 SEXTON, A. HTJMBOLDT. Fuel and Refractory Ma- terials. 8vo, cloth $2.00 SHIELDS, J. E. Notes on Engineering Construction. Embracing Discussions of the Principles involved, and Descriptions of the Material employed in Tunnelling, Bridging, Canal and Road Building, etc. 12mo, cloth $1.50 SHOCK, WM. H. Steam Boilers, Their Design, Con- struction and Management. 4to, half morocco $15.00 SHREVE, S. H. A Treatise on the Strength of Bridges and Roofs. Comprising the determination of Algebraic formulas for strains in Horizontal, Inclined or Rafter, Triangular, Bowstring, Lenticular, and other Trusses, from fixed and moving loads, with practical applications, and examples, for the use of Students and Engineers. 87 woodcut illus. Fourth edition. 8vo, cloth $3.50 SHTJNK, W. F. The Field Engineer. A Handy Book f of practice in the Survey, Location, and Truck-work of Railroads, con- taining a large collection of Rules and Tables, original and selected, applicable to both the Standard and Narrow Gauge, and prepared with special reference to the wants of the young Engineer. Eleventh edition, revised and enlarged. 12mo, morocco, tucks $2.50 SIMMS, F. W. A Treatise on the Principles and Prac- tice of Levelling. Showing its application to purposes of Railway Engineering, and the Construction of Roads, etc. Revised and cor- rected, with the addition of Mr. Laws' Practical Examples for setting oat Railway Curves. Illustrated. 8vo, cloth, $2.50 28 D. VAN NOSTRAND COMPANY'S SIMMS, W. F. Practical Tunnelling. Fourth edition, revised and greatly extended. With additional chapters illustrating recent practice by D. Kinnear Olark. With 36 plates and other illus- trations. Imperial 8vo, cloth $12 00 SLATER, J. W. Sewage Treatment, Purification, and Utilization. A Practical Manual for the Use of Corporations, Local Boards, Medical Officers of Health, Inspectors of Nuisances, Chem- ists, Manufacturers, Riparian Owners, Engineers, and Rate-payers. 12mo, cloth $2.25 SMITH, ISAAC W., C.E. The Theory of Deflections and of Latitudes and Departures. With special applications to Curvilinear Surveys, for Alignments of Railway Tracks. Illustrated. 16mo, morocco, tucks $3.00 - GTJSTAVUS W. Notes on Life Insurance. The- oretical and Practical. Third edition, revised and enlarged. 8vo, cloth , $2.00 SNELL, ALBION T. Electric Motive Power: The Transmission and Distribution of Electric Power by Continuous and Alternate Currents. With a Section on the Applications of Electricity to Mining Work. 8vo., cloth, illustrated $4.00 SPEYERS, CLARENCE L. Text Book of Physical Chemistry. 8vo, cloth $2.25 STAHL, A. W., and A. T. WOODS. Elementary Me- chanism. A Text-Book for Students of Mechanical Engineering. Fourth edition, enlarged. 12mo, cloth $2.00 STALEY, CADY, and PIERSON, GEO. S. The Separ- ate System of Sewerage : its Theory and Construction. Second edi- tion, revised. 8vo, cloth. With maps, plates and illustrations. .$3.00 STEVENSON, DAVID, F.R.S.N. The Principles and Practice of Canal and River Engineering. Revised by his sons David Alan Stevenson, B. Sc., F.R.S.E., and Charles Alexander Stevenson, B. Sc., F.R.S.E., Civil Engineer. Third edition, with 17 plates. 8vo, cloth $10.00 The Design and Construction of Harbors. A Treat- ise on Maritime Engineering. Third edition, with 24 plates. 8vo, cloth $9.00 STEWART, R. W. A Text Book of Light. Adapted to the Requirements of the Intermediate Science and Preliminary Scientific Examinations of the University of London, and also for General Use. Numerous Diagrams and Examples. 12mo, cloth, $1.00 A Text Book of Heat. Illustrated. 8vo, cloth... .$1.00 A Text-Book of Magnetism and Electricity. 160 Ulus. and Numerous Examples. 12mo, cloth $1.00 SCIENTIFIC PUBLICATIONS. 29 STILES, AMOS. Tables for Field Engineers. Designed for use in the field. Tables containing all the functions of a one degree curve, from which a corresponding one can be found for any required degree. Also, Tables of Natural Sines and Tangents. 12mo, morocco, tucks $2.00 STILLMAN, PAUL. Steam-Engine Indicator and the Improved Manometer Steam and Vacuum Gauges ; their Utility and Application. New edition. 12mo, flexible cloth $1.00 STONE, General ROY. New Roads and Road Laws in the United States. 200 pages, with numerous illustrations. 12mo, cloth $1.00 STONEY, B. D. The Theory of Stresses in Girders and Similar Structures. With observations on the application of Theory to Practice, and Tables of Strength, and other properties of Materials. New revised edition, with numerous additions on Graphic Statics, Pillars, Steel, Wind Pressure, Oscillating Stresses, Working Loads, Eiveting, Strength and Tests of Materials. 8vo, 777 pages, 143 illustrations, and 5 folding plates $12.50 STUART, C. B. TJ. S. N. Lives and Works of Civil and Military Engineers of America. With 10 steel-plate engravings. 8vo, cloth $5.00 The Naval Dry Docks of the United States. Illustrated with 24 fine Engravings on Steel. Fourth edition. 4to, cloth $6.00 SWEET, S. H. Special Report on Coal, showing its Distribution, Classification, and Costs delivered over different routes to various points in the State of New York and the principal cities on the Atlantic Coast. With maps. 8vo, cloth $3.00 SWINTON, ALAN A. CAMPBELL. The Elementary Principle of Electric Lighting. Illustrated. 12mo, cloth 60 TEMPLETON, WM. The Practical Mechanic's Work- shop Companion. Comprising a great variety of the most useful rules and formulae in Mechanical Science, with numerous tables of practical data and calculated results facilitating mechanical operations. Eevised and enlarged by W. S. Hutton. 12mo, morocco .$2.00 THOM, CHAS., and WILLIS H. JONES. Telegraphic Connections: embracing Recent Methods in Quadruplex Telegraphy. Oblong, 8vo, cloth. 20 full page plates, some colored $1.50 THOMPSON, EDWARD P., M. E. How to M!ake In- ventions ; or, Inventing as a Science and an Art. A Practical Guide for Inventors. Second edition. 8vo, boards. $1.25 30 D. VAN NOSTKAND COMPANY'S THOMPSON, EDWARD P., M. E. Roentgen Kays and Phenomena of the Anode and Cathode. Principles, Applications and Theories. For Students, Teachers, Physicians, Photographers, Electricians and others. Assisted by Louis M. Pignolet, N. D. C. Hodges, and Ludwig Gutmann, E. E. With a Chapter on Generali- zations, Arguments, Theories, Kindred hadiations and Phenomena. By Professor Wm. Anthony. 8vo, cloth. 50 Diagrams, 40 Half tones $1.50 TODD, JOHN and W. B. WHALL. Practical Seaman- ship for Use in the Merchant Service : Including all ordinary sub- jects; also Steam Seamanship, Wreck Lifting, Avoiding Collision, Wire Splicing, Displacement, and everything necessary to be known by seamen of the present day. Second edition, with 247 illustrations and diagrams. 8vo, cloth . . $8. 40 TOOTHED GEARING. A Practical Hand-Book for Offices and Workshops. By a Foreman Patternmaker. 184 Illustra- tions. 12mo, cloth $2. 25 TRATMAN, E. E. RUSSELL. Railway Track and Track-Work. With over two hundred illustrations. 8vo, cloth. $3. 00 TREVERT, EDWARD. How to build Dynamo-Electric Machinery, embracing Theory Designing and Construction of Dy- namos and Motors. With appendices on Field Magnet and Armature Winding, Management of Dynamos and Motors, and Useful Tables of Wire Gauges. Illustrated. 8vo, cloth $2.50 Electricity and its Recent Applications. A Practi- cal Treatise for Students and Amateurs, with an Illustrated Dictionary of Electrical Terms and Phrases. Illustrated. 12mo, cloth . . $2. 00 TUCKER, Dr. J. H. A Manual of Sugar Analysis, in- cluding the Applications in General of Analytical Methods to the Sugar Industry. With an Introduction on the Chemistry of Cane Sugar, Dextrose, Levulose, and Milk Sugar. 8vo, cloth, illus- trated $3.50 TUMLIRZ, DR. O. Potential and its Application to the Explanation of Electric Phenomena, Popularly Treated. Trans- lated from the German by D. Eobertson. 111. 12mo, cloth $1.25 TUNNER, P. A. Treatise on Roll-Turning for the Manufacture of Iron. Translated and adapted by John B. Pearse. of the Pennsylvania Steel Works, with numerous engravings, woodcuts. 8vo, cloth, with folio atlas of plates $10.00 URQUHART, J. W. Electric Light Fitting. Embody- ing Practical Notes on Installation Management. A Hand-book for Working Electrical Engineers with numerous illustrations. 12mo, cloth $2.00 Electro-Plating. A Practical Hand Book on the Deposition of Copper, Silver, Nickel, Gold, Brass, Aluminium, Plat- ininum, etc. Third edition. 12mo $2.00 SCIENTIFIC PUBLICATIONS. 31 URQUHART, J. W. Electrotyping. A Practical Man- ual forming a New and Systematic Guide to the Reproduction and Multiplication of Printing Surfaces, etc. 12mo $2.00 Dynamo Construction : a Practical Hand-Book for the Use of Engineer Constructors and Electricians in Charge, em- bracing Frame Work Building, Field Magnet and Armature Winding and Grouping, Compounding, etc.. with Examples of Leading Eng- lish, American and Continental Dynamos and Motors, with numerous illustrations. 12mo, cloth $3.00 Electric Ship Lighting. A Hand-Book on the Practical Fitting and Running of Ship's Electrical Plant. For the Use of Ship Owners and Builders, Marine Electricians and Sea Going Engineers-in-Charge. Numerous illustrations. 12mo, cloth $3.00 UNIVERSAL TELEGRAPH CIPHER CODE. Ar- ranged for General Correspondence. 12mo, cloth $1.00 VAN HEURCK, Dr. HENRI. The Microscope, Its Construction and Management, including Technique, Photo-Micro- graphy and the Past and Future of the Microscope. English edition re-edited and augmented by the Author from the fourth French edition, and translated by Wynne E. Baxter, F.R.M.S. 3 Plates and upwards of 250 Illustrations. Imperial 8vo, cloth $7.00 VAN NOSTRAND'S Engineering Magazine. Com- plete sets, 1869 to 1 886 inclusive. Complete sets, 35 vols. , in cloth $60.00 Complete sets, 35 vols. , in half morocco 100.00 VAN WAGENEN, T. F. Manual of Hydraulic Mining. For the Use of the Practical Miner. Revised and enlarged edition. 18mo, cloth $1.00 WALKER, W. H. Screw Propulsion. Notes on Screw Propulsion, its Rise and History. 8vo, cloth 75 WALKER, SYDNEY F. Electrical Engineering in Our Homes and Workshops. A Practical Treatise on Auxiliary Elec- trical Apparatus. Third edition, revised, with numerous illustra- tions $2.00 - Electric Lighting for Marine Engineers, or How to Light a Ship by the Electric Light and How to Keep the Apparatus in Order. 103 illustrations. 8vo, cloth. Second edition $2.00 WALLIS-TAYLER, A. J. Modern Cycles, A Practi- cal Handbook on Their Construction and Repair. With 300 illustra- tions. 8vo, cloth $4.00 Motor Cars, or Power Carriages for Common Roads. 8vo, cloth, with numerous illustrations $1.80 Bearings and Lubrication. A Handbook for every user of Machinery. 8vo, cloth, fully illustrated $1.50 32 D. VAN NOSTBAND COMPANY'S WALLJS-TAYLER, A. J. Refrigerating and Ice- Making Machinery. A Descriptive Treatise for the use of persons employing refrigerating and ice-making installations and others. 8vo, cloth, illustrated $3.00 Sugar Machinery. A Descriptive Treatise, De- voted to the Machinery and Apparatus used in the Manufacture of Cane and Beet Sugars. 12mo, cloth, ill $2.00 WANKLYN, J. A. A Practical Treatise on the Exam- ination of Milk and its Derivatives, Cream, Butter, and Cheese. 12mo, cloth $1.00 - Water Analysis. A Practical Treatise on the Ex- amination of Potable Water. Tenth edition. 12mo, cloth $2.00 WANSBROUGH, WM. D. The A. B. C. of the Differ- ential Calculus. 12mo, cloth $1.50 WARD, J. H. Steam for the Million. A Popular Treat- ise on Steam, and its application to the Useful Arts, especially to Navigation. 8vo, cloth $1.00 WARING, GEO. E., Jr. Sewerage and Land Drainage. Illustrated with wood-cuts in the text, and full-page and folding plates. Quarto. Cloth. Third edition $6.00 Modern Methods of Sewage Disposal for Towns, Public Institutions and- Isolated Houses. Second edition, revised and enlarged. 260 pages. Illustrated, cloth $2.00 How to Drain a House. Practical Information for Householders. New and enlarged edition.] 12mo, cloth $1.25 WATSON, E. P. Small Engines and Boilers. A man- ual of Concise and Specific Directions for the Construction of Small Steam Engines and Boilers of Modern Types from five Horse-power down to model sizes. 12mo, cloth. Illustrated with Numerous Diagrams and Half Tone Cuts. New York, 1899 $1.25 WATT, ALEXANDER. Electro-Deposition. A Prac- tical Treatise on the Electrolysis of Gold, Silver, Copper, Nickel, and other Metals, with Descriptions of Voltaic Batteries, Magneto and Dynamo-Electric Machines, Thermopiles, and of the Materials and Processes used in every Department of the Art, and several chapters on Electro-Metallurgy. With numerous illustrations. Third edition, revised and corrected. Crown, 8vo, 568 pages , .... $3. 50 Electro-Metallurgy Practically Treated. Tenth edition, considerably enlarged. 12mo, cl $1.00 The Art of Soap-Making. A Practical Handbook of the Manufacture of Hard and Soft Soaps, Toilet Soaps, &c. In- cluding many New Processes, and a Chapter on the Eecovery of Glycerine from Waste Lyes. With illustrations. Fourth edition, revised and enlarged. 8vo $3. 00 SCIENTIFIC PUBLICATIONS. 33 WATT, ALEXANDER. The Art of Leather Manufact- ure. Being a Practical Handbook, in which the Operations of Tan- ning, Currying, and Leather Dressing are Fully Described, and the Principles of Tanning Explainad, and many Eecent Processes Intro- duced. With numerous illustrations. Second edition. 8vo, cl.$4.00 WEALE, JOHN. A Dictionary of Terms Used in Architecture, Building, Engineering, Mining, Metallurgy, Arehaelogy, the Fine Arts, etc. , with explanatory observations connected with applied Science and Art. fifth edition, revised and corrected. 1.2mo, cloth $2.50 Weale's Rudimentary Scientific Series, [see list, p. 39. J WEBB, HERBERT LAWS. A Practical Guide to the Testing of Insulated Wires and Cables. Illustrated. 12mo, cloth $1.00 -The Telephone Hand Book. 128 illustrations. 146 pages. 16mo. , cloth $1. 00 WEEKES, R. W. The Design of Alternate Current Transformers. Illustrated. 12mo, cloth .$1.00 WEISBACH, JULIUS. A Manual of Theoretical Mechanics. Eighth American edition. Translated from the fourth augmented and improved German edition, with an Introduction to the Calculus by Eckley B. Coxe, A.M., Mining Engineer. 1,100 pages, and 902 woodcut illustrations. 8vo, cloth $6.00 Sheep. 7.50 WESTON, EDMUND B. Tables Showing Loss of Head Due to Friction of Water in Pipes. Second edition. 12mo, cloth $1.50 WEYMOUTH, F. MARTEN. Drum Armatures and Commutators. (Theory and Practice.) A complete Treatise on the Theory and Construction of Drum Winding, and of commutators for closed-coil armatures, together with a full resume of some of the prin- cipal points involved in their design, and an exposition of armature re-actions and sparking. 8vo, cloth $3.00 WEYRAUCH, J. J. Strength and Calculations of Dimensions of Iron and Steel Construction, with reference to the Latest Experiments. 12mo, cloth, plates $1.00 WHEELER, Prof, J. B. Art of War. A Course of Instruction in the Elements of the Art and Science of War, for the Use of the Cadets of the United States Military Academy, West Point, N. Y. 12mo, cloth .$1.75 Field Fortifications. The Elements of Field Fortifications, for the Use of the Cadets of the United States Military Academy, West Point, N. Y. 12mo SI. 75 34 D. VAN NOSTRAND GO'S SCIENTIFIC PUBLICATIONS. WHIPPLE, S., C. E. An Elementary and Practical Treatise on Bridge Building. 8vo, cloth $3.00 WILKINSON, H. D. Submarine Cable-Laying, Re- pairing and Testing. 8vo, clotli $4.00 WILLIAMSON, R. S. On the Use of the Barometer on Surveys and Reconnoissances. Part I. Meteorology in its Connection with Hypsometry. Part II. Barometric Hypsometry. With Illus- trative tables and engravings. 4to, cloth $15.00 Practical Tables in Meteorology and Hypsometry, in connection with the use of the Barometer. 4to, cloth $2.50 WILSON, GEO. Inorganic Chemistry, with New No- tation. Bevised and enlarged by H. G. Madan. New edition. 12mo, cloth $2.00 WILL YOUNG, E. G., and W. M. STONE, Ph. D. Elec- trical Instruments and Measurements. 12mo, cl. 2 vols.. . .In Press. WOODBURY, D. V. Treatise on the Various Elements of Stability in the Weil-Proportioned Arch. 8vo, half morocco.. $4. 00 WRIGHT, T. W. A Treatise on the Adjustment of Observations. With applications to Geodetic Work, and other Meas- ures of Precision. 8vo, cloth $4.00 Elements of Mechanics ; including Kinematics, Kinetics and Statics. With applications. 8vo, cloth $2.50 WYLIE, CLAUDE. Iron and Steel Founding. Illus- trated with 39 diagrams. Second edition, revised and enlarged. 8vo, cloth , $2.0-, WYNKOOP, RICHARD. Vessels and Voyages, as Regulated by Federal Statutes and Treasury Instructions and Decis- ions. 8vo, cloth , $2.00 YOUNG, J. ELTON. Electrical Testing for Telegraph Engineers, with Appendices consisting of Tables. 8vo, cloth, illus- trated $4.00 YOUNG SEAMAN'S MANUAL. Compiled from Various Authorities, and Illustrated with Numerous Original and Select Designs, for the Use of the United States Training Ships and the Marine Schools. 8vo, half roan $3.00 Catalogue of the Van Nostrand Science Series. *T*HE I ' are put up in a uniform, neat, and attractive form. -1 boards. Price jo cents per volume. The subjects are of an eminently scientific character, and embrace a wide range of topics, and are amply illustrated when the subject demands. No. i. CHIMNEYS FOR FURNACES AND STEAM BOILERS. By R. Armstrong, C.E. Third American edition, revised and partly rewritten, with an appendix on Theory of Chimney Draught, by F. E. Idell, M.E. No. 2. STEAM-BOILER EXPLOSIONS. By Zerah Colburn. Nw edition, revised by Prof. R. H. Thurston. No. 3. PRACTICAL DESIGNING OF RETAINING- WALLS. By Arthu W. Cain. Arthur Jacob, A.B. Second edition, revised, with additions by Prof. No. 4. PROPORTIONS OF PINS USED IN BRIDGES. Second edition, with appendix. By Charles E. Bender, C.E. No. 5. VENTILATION OF BUILDINGS. By W. F. Butler. Second edition, re-edited and enlarged by James L. Greenleaf, C.E. No. 6. ON THE DESIGNING AND CONSTRUCTION OF STORAGE RESERVOIRS. By Arthur Jacob, A.B. Second edition, revised, with additions by E. Sherman Gould. No. 7. SURCHARGED AND DIFFERENT FORMS OF RE- TAINING-WALLS. By James S. Tate, C.E. No. 8. A TREATISE ON THE COMPOUND ENGINE. By John Turnbull, jim. Second edition, revised by Prof. S. W. Robinson. No. 9. A TREATISE ON FUEL. By Arthur V. Abbott, C. E. Founded on the original treatise of C. William Siemens, D.C.L. No. 10. COMPOUND ENGINES. Translated from the French of A. Mallet. Second edition, revised, with Results of American Practice, by Richard II. Buel, C.E. No. ii. THEORY OF ARCHES. By Prof. W. Allan. No. 12. A THEORY OF VOUSSOIR ARCHES. By Prof. W. E. Cain. Second edition, revised and enlarged. Illustrated. No 13. GASES MET WITH IN COAL-MINES. By J. J. Atkinson. Third edition, icvised and enlarged by Edward H Williams, jun. D. VAN NO STRAND COMPANY'S No. 14. FRICTION OF AIR IN MINES. By J. J. Atkinson, No. 15. SKEW ARCHES. By Prof. E. W. Hyde, C.E. Illustrated No. 16. A GRAPHIC METHOD FOR SOLVING CERTAIN ALGEBRAIC EQUATIONS. By Prof. George L. Vose. No. 17. WATER AND WATER-SUPPLY. By Prof. W. H. Corfield of the University College, London. No. 18. SEWERAGE AND SEWAGE PURIFICATION. By M. N. Baker, Assoc. Ed. Engineering News. No. 19. STRENGTH OF BEAMS UNDER TRANSVERSE LOADS. By Prof. W. Allan, author of "Theory of Arches." No. 20. BRIDGE AND TUNNEL CENTRES. By John B. Me- Master, C.E. No. 21. SAFETY VALVES. By Richard H. Buel, C.E. Second edition. No 22. HIGH MASONRY DAMS. By E. Sherman Gould, C.E. No. 23. THE FATIGUE OF METALS UNDER REPEATED STRAINS. With Various Tables ot Results and Experiments. From the German of Prof. Ludwig Spangenburgh, with a Preface by S. li. Shreve, A.M. No. 24. A PRACTICAL TREATISE ON THE TEETH OF WHEELS. By Prof. S. W. Robinson. Second edition, revised. No. 25. ON THE THEORY AND CALCULATION OF CAN- TILEVER BRIDGES. By R. M. Wilcox, Ph.B. No. 26. PRACTICAL TREATISE ON THE PROPERTIES Or CONTINUOUS BRIDGES. By Charles Bender, C.E. No. 27. ON BOILER INCRUSTATION AND CORROSION By F. J. Rowan. New edition, revised and partly rewritten by F. L. Idell, M. E. No. 28. TRANSMISSION OF POWER BY WIRE ROPES By Albert W. Stahl, U.S.N. Second edition. No. 29. STEAM INJECTORS. Translated from the French oi M. Leon Pochet. No. 30. TERRESTRIAL MAGNETISM, AND THE MAGNET- ISM OF IRON VESSELS. By Prof. Fairman Rogers. .No. 31. THE SANITARY CONDITION OF DWELLING- HOUSES IN TOWN AND COUNTRY. By George E. Waring, jun, No. 32. CABLE-MAKING FOR SUSPENSION BRIDGES. By W. Hildenbrand, C.E. No. 33. MECHANICS OF VENTILATION. By George W. Rafter, C.E. New edition (1895), revised by author. No. 34. FOUNDATIONS. By Prof. Jules Gaudard, C.E. Translated from the French. No. 35. THE ANEROID BAROMETER : ITS CONSTRUC- TION AND USE. Compiled by George W. Plympton. Fourth edition No. 36. MATTER AND MOTION. By J. Clerk Maxwell, M.A. Second American edition. SCIENCE SERIES. Mo. 37. GEOGRAPHICAL SURVEYING: ITS USES, METH- ODS, AND RESULTS. By Frank De Yeaux Carpenter, C.E. No. 38. MAXIMUM STRESSES IN FRAMED BRIDGES. By Prof. William Cain, A.M., C.E. New and revised edition. No. 39. A HANDBOOK OF THE ELECTRO-MAGNETIC TELEGRAPH. By A. E. Loring. No. 40. TRANSMISSION OF POWER BY COMPRESSED AIR. By Robert Zahner, M.E. Second edition. No. 41. STRENGTH OF MATERIALS. By William Kent, C.E., Assoc. Ed. Engineering News. No. 42. VOUSSOIR ARCHES APPLIED TO STONE BRIDGES, TUNNELS, CULVERTS, AND DOMES. By Prof. William Cain. No. 43. WAVE AND VORTEX MOTION. By Dr. Thomas Craig of Johns Hopkins University. No. 44. TURBINE WHEELS. By Prof. W. P. Trowbridge, Columbia College. Second edition. No. 45. THERMODYNAMICS. By Prof. H. T. Eddy, University of Cincinnati. No. 46. ICE-MAKING MACHINES. New edition, revised and en- larged by Prof. J. E. Denton. From the French of M. Le Doux. No. 47. LINKAGES; THE DIFFERENT FORMS AND USES OF ARTICULATED LINKS. By J. D. C. de Roos. No. 48. THEORY OF SOLID AND BRACED ARCHES. By William Cain, C.E. No. 49. ON THE MOTION OF A SOLID IN A FLUID. By Thomas Craig, Ph.D. No. 50. DWELLING-HOUSES: THEIR SANITARY CON- STRUCTION AND ARRANGEMENTS. By Prof. W. H. Corfield. No. 51. THE TELESCOPE : ITS CONSTRUCTION, ETC. By Thomas Nolan. No. 52. IMAGINARY QUANTITIES. Translated from the French of . . M. Argand. By Prof. Hardy. No. 53. INDUCTION COILS : HOW MADE AND HOW USED. Fifth edition. No. 54. KINEMATICS OF MACHINERY. By Prof. Kennedy. With an introduction by Prof. R. H. Thurston. No. 55. SEWER GASES : THEIR NATURE AND ORIGIN. By A. de Varona. No. 56. THE ACTUAL LATERAL PRESSURE OF EARTH- WORK. By Benjamin Baker, M. Inst. C.E. No. 57. INCANDESCENT ELECTRIC LIGHTING. A Practical Description of the Edison System. By L. H. Latimer, to which is added the Design and Operation of Incandescent Stations, by C. J. Field, and the Maximum Efficiency of Incandescent Lamps, by John W. Ho well. N - 5 -/ r | I !Fo VENTILATION OF COAL-MINES. By W. Fairlev. M.llr , F.S.S. D. VAN NO STRAND COMPANY'S No. 59. RAILROAD ECONOMICS; OR, NOTES, WITH COM- MENTS. By S. W. Robinson, C.E. No. 60. STRENGTH OF WROUGHT-IRON BRIDGE MEM- BERS. By S. W. Robinson, C E. No. 61. POTABLE WATER AND METHODS OF DETECT- ING IMPURITIES. By M. N. Baker, Ph.B. No. 62. THE THEORY OF THE GAS-ENGiNE. By Dugald Clerk. Second edition. With additional matter. Edited by F. E. Idell, M.E. No. 63. HOUSE DRAINAGE AND SANITARY PLUMBING. By W. P. Gerhard. Seventh edition, revised. No. 64. ELECTRO-MAGNETS. ByTh.du Moncel. 2d revised edition. No. 65. POCKET LOGARkTHMS TO FOUR PLACES OF DECI- MALS. No. 66. DYNAMO-ELECTRIC MACHINERY. By S. P. Thompson, With notes by F. L. Pope. Third edition. No. 67. HYDRAULIC TABLES BASED ON " KUTTER'S FORMULA." By P. J. Flynn. No. 68. STEAM-HEATING. By Robert Briggs. Second edition, revised, with additions by A. R. Wolff. No. 69. CHEMICAL PROBLEMS. By Prof. J. C. Foye. Fourth edition, revised and enlarged. No. 70. EXPLOSIVE MATERIALS. The Phenomena and Theories of Explosion, and the Classification, Constitution and Preparation of Explosives. By First Lieut. John P. Wisser, U.S.A. No. 71. DYNAMIC ELECTRICITY. By John Hopkinson, J. A. Schoolbred, and^R. E. Day. No. 72. TOPOGRAPHICAL SURVEYING. By George J. Specht, Prof. A. S. Hardy, John B. McMaster, and H. F. Walling. No. 73. SYMBOLIC ALGEBRA; OR, THE ALGEBRA OF ALGEBRAIC NUMBERS. By Prof. W. Cain. No. 74. TESTING MACHINES : THEIR HISTORY, CON- STRUCTION, AND USE. By Arthur V. Abbott. No. 75. RECENT PROGRESS IN DYNAMO-ELECTRIC MA- CHINES. Being a Supplement to Dynamo-Electric Machinery. By Prof. Sylvanus P. Thompson. No. 76. MODERN REPRODUCTIVE GRAPHIC PROCESSES. By Lieut. James S. Pettit, U.S.A. No. 77. STADIA SURVEYING. The Theory ot Stadia Measurements. By Arthur Winslow. No. 78. THE STEAM-ENGINE INDICATOR, AND ITS USE By W. B. Le Van. No. 79. THE FIGURE OF THE EARTH. By Frank C. Roberts, C.E. No. 80. HEALTH* FOUNDATIONS FOR HOUSES. By Glea Brown. SCIENCE SERIES. No. 81. WATER METERS : COMPARATIVE TESTS OF ACCURACY, DELIVERY, ETC. Distinctive features of the Worth- ington, Kennedy, Siemens, and Hesse meters. By Ross E. Browne. No. 82. THE PRESERVATION OF TIMBER BY THE USE OF ANTISEPTICS. By Samuel Bagster Boulton, C.E. No. 83. MECHANICAL INTEGRATORS. By Prof. Henry S. H. Shaw, C.E. No. 84. FLOW OF WATER IN OPEN CHANNELS, PIPES, CONDUITS, SEWERS, ETC. With Tables. By P. J. Flynn, C.E. No. 85 THE LUMINIFEROUS AETHER. By Prof, de Volson Wood. No. 86. HAND-BOOK OF MINERALOGY; DETERMINATION AND DESCRIPTION OF MINERALS FOUND IN THE UNITED STATES. By Prof. J. C. Foye. No. 87. TREATISE ON THE THEORY OF THE CON- STRUCTION OF HELICOIDAL OBLIQUE ARCHES. By John L. Culley, C.E. No. 88. BEAMS AND GIRDERS. Practical Formulas for their Re- sistance. By P. H. Philbrick. No. 89. MODERN GUN-COTTON: ITS MANUFACTURE, PROPERTIES, AND ANALYSIS. By Lieut. John P. Wisser, U.S.A. No. 90. ROTARY MOTION, AS APPLIED TO THE GYRO- SCOPE. By Gen. J. G. Barnard. No. 91. LEVELING: BAROMETRIC, TRIGONOMETRIC, AND SPIRIT. By Prof. I. O. Baker. No. 92. PETROLEUM : ITS PRODUCTION AND USE. By Boverton Redwood, F.I.C., F.C.S. No. 93. RECENT PRACTICE IN THE SANITARY DRAIN- AGE OF BUILDINGS. With Memoranda on the Cost of Plumbing Work. Second edition, revised. By William Paul Gerhard, C. E. No. 94. THE TREATMENT OF SEWAGE. By Dr. C. Meymott Tidy. No. 95. PLATE GIRDER CONSTRUCTION. By Isami Hiroi, C.E. Second edition, revised and enlarged. Plates and Illustrations. No. 96. ALTERNATE CURRENT MACHINERY. By Gisbert Kapp, Assoc. M. Inst., C.E. No. 97. THE DISPOSAL OF HOUSEHOLD WASTE. By W, Paul Gerhard, Sanitary Engineer. No. 98. PRACTICAL DYNAMO-BUILDING FOR AMATEURS. HOW TO WIND FOR ANY OUTPUT. By Frederick Walker. Fully illustrated. No. 99. TRIPLE-EXPANSION ENGINES AND ENGINE TRIALS. By Prof. Osborne Reynolds. Edited, with notes, etc., by F. E. Well. M. E. SCIENCE SERIES. No. 100. HOW TO BECOME AN ENGINEER ; OH THE THEORETICAL AND PRACTICAL TRAINING NECESSARY IN FITTING FOR THE DUTIES OF THE CIVIL ENGINEER. The Opinions of Eminent Authorities, and the Course of Study in the Technical Schools. By Geo. W. Plympton, Am. Soc. C.E. No. 101. THE SEXTANT AND OTHER REFLECTING MATHEMATICAL INSTRUMENTS. With Practical Suggestions and Wrinkles on their Errors, Adjustments, and Use. With thirty- three illustrations. By F. R. Brainard, U.S.N. No. 102. THE GALVANIC CIRCUIT INVESTIGATED MATHEMATICALLY. By Dr. G. S. Ohm, -Berlin, 1827. Translated by William Francis. W.-li Preface and Notes by the Editor, Thomas D. Lockwood, M.I.E.E. No. 103. THE MICROSCOPICAL EXAMINATION OF POTA- BLE WATER. With Diagrams. By Geo. W. Rafter. No. 104. VAN NOSTRAND'S TABLE-BOOK FOR CIVIL AND MECHANICAL ENGINEERS. Compiled by Geo. W. Plympton, C.E, No. 105. DETERMINANTS, AN INTRODUCTION TO THE STUDY OF. With examples. By Prof. G. A. Miller. No. 106. TRANSMISSION BY AIR-POWER. Illustrated. By Prof. A. B. W. Kennedy and W. C. Unwin. No. 107. A GRAPHICAL METHOD FOR SWING-BRIDGES. A Rational and Easy Graphical Analysis of the Stresses in Ordinary Swing-Bridges. With an Introduction on the General Theory of Graphi- cal Statics. 4 Plates. By Benjamin F. LaRue, C.E. No. 108. A FRENCH METHOD FOR OBTAINING SLIDE: VALVE DIAGRAMS. 8 Folding Plates. By Lloyd Bankson, B.S., Assist. Naval Constructor, U.S.N. No. 109. THE MEASUREMENT OF ELECTRIC CURRENTS. ELECTRICAL MEASURING INSTRUMENTS. By Jas. Swinburne. METERS FOR ELECTRICAL ENERGY. By C. H. Wordingham. Edited by' T. Commerford Martin. Illustrated. No. no. TRANSITION CURVES. A Field Book for Engineers, containing Rules and Tables for laying out Transition Curves. By Walter G. Fox. No. in. GAS-LIGHTING AND GAS-FITTING, including Specifica, tions and Rules for Gas Piping, Notes on the Advantages of Gas for Cooking and Heating, and useful Hints to Gas Consumers. Second edition, rewritten and enlarged. By Wm. Paul Gerhard. No. 112. A PRIMER ON THE CALCULUS. By E. Sherman Gould, C.E. No. 113. PHYSICAL PROBLEMS AND THEIR SOLUTION. By A. Bourgougnon, formerly Assistant at Bellevue Hospital. No. 114. MANUAL OF THE SLIDE RULE. By F. A. Halsey of the American Machinist. THIS BOOK IS DUE ON THE LAST DATE STAMPED BELOW INITIAL FlNE~OF 25 CENTS OVERDUE. OCT 1933 10 1836 LD 21-50m-l, Pt 79 01 '